Jul

26

A mass sacrifice of children and camelids at the Huanchaquito-Las Llamas site, Moche Valley, Peru

Here we report the results of excavation and interdisciplinary study of the largest child and camelid sacrifice known from the New World. Stratigraphy, associated artifacts, and radiocarbon dating indicate that it was a single mass killing of more than 140 children and over 200 camelids directed by the Chimú state, c. AD 1450. Preliminary DNA analysis indicates that both boys and girls were chosen for sacrifice. Variability in forms of cranial modification (head shaping) and stable isotope analysis of carbon and nitrogen suggest that the children were a heterogeneous sample drawn from multiple regions and ethnic groups throughout the Chimú state. The Huanchaquito-Las Llamas mass sacrifice opens a new window on a previously unknown sacrificial ritual from fifteenth century northern coastal Peru. While the motivation for such a massive sacrifice is a subject for further research, there is archaeological evidence that it was associated with a climatic event (heavy rainfall and flooding) that could have impacted the economic, political and ideological stability of one of the most powerful states in the New World during the fifteenth century A.D.

Laurel Kenner comments:

In Lessons from History, the Durants write that Peru was a happy socialist state until the arrival of the conquistadors in the 16C.

Bo Keely reports:

Iquitos, Peru at the headwaters of the Amazon Rio is the only of two places I've lived in the past 20 years. The other is here in Slab City. I had a trip planned to Peru this month but got a desert skin infection that the jungle would have ravaged. As such, i've lived in the Peruvian Amazon a half-dozen times for months at a stint, all in the jungle hiking and hitchhiking banana boats. The proposed postponed trip was to hitch the rios again doing magic tricks for the natives in putting together a photo-essay. The Peruvian Amazon is my haunt because the people operate very low on the brainstem. Cannibalism and malaria make them perhaps the greatest evolved and toughest humans on the planet. Put succinctly, if one is invited to dinner make sure the host isn't licking his chops. I'll go back, and escape again with magic.

Asindu Drileba is concerned:

Put succinctly, if one is invited to dinner make sure the host isn't licking his chops. I'll go back, and escape again with magic.

You have unlocked a whole new level to what I consider a set of risks people take. Please don't do that again.

Jul

24

‘Greatest Bubble’ Nearing Its Peak, Says Black Swan Manager

Universa’s Mark Spitznagel, who has made billions from past crashes, sees last hurrah for stocks before severe reckoning

Humbert H. asks:

His job is to make money on black swans, not to predict black swans. What kind of black swan is it if it can be predicted?

Asindu Drileba writes:

Black Swans are relative. If you have tail risk protection it means you are aware of tail risk. If you don't have tail risk protection, the notion of a "surprise" when it happens means you encounter a black swan. So Mark may be speaking form the perspective of those that actually don't think they will encounter a black swan.

Humbert H. responds:

Is there anyone who invests in the magnificent seven and NVDA in particular who isn't aware of their elevated valuations, possible bubble formation, and the risk of a major decline? There's some level of obviousness to warning people of this possibility. It's like he is suddenly preaching "past performance is no guarantee of future results" or "correlation does not equal causation". Is he doing this to help humanity? Someone will make more money and someone will make less money if they act on his warning, and there will be bagholders either way, so humanity will not benefit as a whole.

Asindu Drileba adds:

I think for his case, he is just marketing his fund.

Zubin Al Genubi observes:

Cheap Deep OTM puts are up 45% on a 3% decline showing exponential gearing in place from ATH as a directional trade or as a hedge. Surprisingly unidirectional.

Asindu Drileba expands:

His philosophy is more like that of "insurance" for stocks. I think Uncle Roy also has the same philosophy. I remember his describing portfolio protection akin to having fire insurance for your house. To benefit from fire insurance on your house, you don't need to predict when it will burn down. Just make sure you always have coverage for it. So most of the time, percentage wise, your predictions of having a fire are going to be wrong. He mostly advocates that everyone should have "fire insurance" for your portfolio.

To learn more about Mark's strategy:
1) A section called "The Forest In the Pine Cone" inside his book The Dao of Capital
2) His solution to the "narrow framing" problem
3) How he sizes his positions

Nils Poertner comments:

good to be open minded. that said: the more stories (like this one) we can read in mass financial media (FT, WSJ, etc) - the less likely this is going to play out anytime soon. "get the joke"

Humbert H. writes:

I don't think it makes any difference unless "everybody" has the opposite view of the future from what the market is doing. Every single day multiple people prognosticate both doom and gloom and full steam ahead. Since the motivation for this warning is clearly suspect this is white noise. But if his prediction comes true soon which it obviously has a reasonable chance of doing he'll be venerated for decades as the great prophet. This guy is clearly a disciple of Taleb, and they even collaborated in the past. Victor's take would be interesting.

Jul

21

From the original version of the Daily Spec site and worth a review:

The Speculator's Reading List

Jeff Watson writes:

Being There is a movie adapted from Jerzy Kosinski’s book about a gardener who took Washington DC by storm. His name was Chance and he could not read or write, but the public thought he was a genius. He ultimately became the President of the United States. The book should also be on the list.

Stefan Jovanovich suggests:

Shattered Sword

Asindu Drileba offers:

- Risk Savvy by Gerd Gigerenzer. How to cut your cancer risk by 50%, how to beat Nobel Prize portfolio strategies, why certainty is an illusion. I think everyone can benefit at least one thing from reading this book. It doesn't matter if you're a spec or not.

- The Visual Display of Quantitative Information (everything by Edward R. Tufte is worth reading)

- Adam Curtis documentaries. He has dedicated his life talking about "Power", mostly the relationship between Markets, Politics, Science, Religion & Philosophy. He informed alot of my thinking about the relationship between those. An incomplete assortment.

- Zurich Axioms. This was recommended by on a podcast. I think it was Larry Williams (but I am not sure). It's a very good book of aphorisms, useful to get your psychology right.

- This is the Road to Stock Market Success (1944). Recommend by Vic. I also find the book very instrumental in developing a psychological edge.

Zubin Al Genubi recommends:

Conrad, J, Heart of Darkness. A river trip into Africa loses grip.

Khilav Majmudar agrees:

Loved Heart of Darkness. Conrad's writing is hypnotic.

Humbert H. adds:

Heart of Darkness is kind of similar to Kafka’s writing in that it’s mysterious and unusual, and nobody knows what it’s really about after reading it. It was famously an inspiration for the movie Apocalypse Now which is arguably even stranger.

Jul

18

Investing in places where property rights are fundamentally not respected just isn't worth it because you can't calculate the risk.

This is completely true. In my country (Uganda) there is a lot of what you may call "arbitrary use of power". If you a Ugandan and are political connected enough, you can screw almost any foreign or local financier by summoning a "presidential directive". Here are some examples within the past 2 - 3 years:

1. Some guys lent money to a financier and used the mosque as collateral. When the mosque defaulted on the loan, the Moslem community appealed for "help" from politicians. The liquidation of the mosque was blocked.

2. In another instance, some foreign financiers lend money to some local "tycoons". When they defaulted on their loans, the courts suddenly declared that the foreign entities did not have enough paperwork to actually lend the local "tycoons" any money. So the "tycoons" didn't have to pay anything back. This didn't happen once, but *thrice* in the last 3 or so years.

3. Last year (2023) people woke up one day to discover that all licenses and permits to export timber were cancelled. So if you invested money into forestry for export, your return on investment was basically marked to zero.

So if you do business in these places:
a) Be well connected politically
b) Plan your exit at least as well as how you enter into business
c) Don't think long term
d) Restrict yourself to a business that is easy to move to neighbouring country.

A group of financiers however that has seen some success seems to be Venture Capitalists. They are well positioned in b) & d). VCs for example will not touch your business if it is not a Delaware C-Corp. This limits how much damage local politics can affect your business. Also, VCs tend to favour software businesses & service businesses over say manufacturing. This makes them easy to expand or ship operations at a whim. Some VC backed Startups for example just have 1 employee per country just for legal & administrative purposes.

Humbert H. writes:

Regarding publicly listed companies in sub-Saharan Africa, I do not think there are many that would fit the requirements of global institutional investors. This fact contributes to the challenges that private equity firms have in finding exits for investee companies.

Many of the listed companies in Sub-Saharan Africa are owned by the public pension funds in the corresponding country. For example, South African pension systems (PIC, GEPF, ESKOM) are significant shareholders in companies listed on the JSE therefore there is limited liquidity.

Every time I go there I am reminded how small the national economies of countries in sub-Saharan Africa are relative to Europe and the US. In 2010 Goldman Sachs published a paper called Lions on the Move which sounded a bullish tone; however, in my view most of what they predicted did not come to fruition.

Sub-Saharan Africa needs to produce higher value products with their raw material and natural resources on the continent as opposed to simply exporting their raw material and natural resources.

H. Humbert comments:

Historically, property rights were respected in Northern Europe and some of their former settlements/colonies, parts of Central Europe, China, Japan, Korea, small parts of Italy, and small enclaves in Southeast Asia and the Iberian Peninsula. These are relatively high-trust societies. Among many problems I have with mass migration is that it always flows from low-trust societies to high-trust societies, which can't be good for the latter. Migration and ever-decreasing penalties for property crimes even in the high-trust societies leave very few counties safe. Switzerland and Japan are likely to last the longest, Germany is doing pretty well although I think it's doomed long term, but the rest of the world is circling the drain. It's possible to reverse trends locally, as recently demonstrated in El Salvador and Argentina, but that's not common. In Russia and Ukraine, if you didn't give bribes or had "blat" (connections) nothing could be done. I don't know the current details, but they're obviously still highly corrupt, as there was no reason for high trust to be established. China (which is a historically high-trust society but corrupted by the Communist dictatorship) and India are complex and have elements of both high-trust and low-trust. If you look at the map of the world, between Russia, Africa, Latin America, and (arguably) a lot of Asia property rights are rare indeed.

Jul

15

Only some people agree, but the power industry believes there may be a demand-supply mismatch from AI data centers. Here are some summary views - from the American Nuclear Society's Nuclear Newswire (April 2024):

Major tech companies see artificial intelligence (AI) as something that will transform their industry, and there is a race to be first. When they look for clean, dependable power 24/7, nuclear clearly stands out as a good match. Constellation [the nation's largest nuclear utility] summarized it best in its recent forecast:
• AI and data center growth will drive power demand.
• Major tech companies are expected to invest $1 trillion in data centers over the next five years.
• In the next five years, consumers and businesses will generate twice as much data as all the data created over the past 10 years.
• AI data center racks could require seven times more power than traditional data center racks.
• Between now and 2030, domestic data center electricity consumption is expected to grow anywhere from 6.5 percent to 7.5 percent (335 terawatt-hours to 390 terawatt-hours).
• In its report, Data Centers 2024 Global Outlook, global real estate services company JLL has said that "AI is driving extreme scale for new developments with requirements now ranging from 300 megawatts (MW) to over 500 MW."

From the IEEE Spectrum (June 2024):

Scientists have predicted that by 2040, almost 50 percent of the world's electric power will be used in computing. What's more, this projection was made before the sudden explosion of generative AI.

From Data Center Dynamics (May 2024):

US utility Dominion expects to connect 15 more data centers to the grid in Virginia over the course of 2024, after connecting 15 facilities last year totaling almost a gigawatt of capacity [1 gigawatt = 1 nuclear plant]. In its most recent earnings presentation this week, the company said it had connected 94 data centers with more than 4GW of capacity in Northern Virginia since 2019. This included 15 data centers totaling 933MW in 2023, and 15 more are due to be connected in 2024. The company didn't include the capacity of those 15 facilities going live this year, and in the earnings call, CEO Robert Blue said he doesn't know how quickly they will ramp up to full capacity.

For those who think new nuclear power is the solution (2024), this is not a quote but a fact: The new Vogtle nuclear power plant took about 20 years to design and build, from concept to commercial operations. This recent construction schedule was set by an experienced nuclear utility that previously built access to transmission on a nuclear site they've owned for decades.

The critical metric is not the overall demand. Data centers' demand sits on the grid 24/7, so generating capacity must be available 24/7. While massive amounts of energy are already oversupplying some US power markets, most new sources originate from part-time wind, solar, and battery assets. Those part-time assets cannot serve the 24/7 load demanded by data centers. Therefore, the critical metric is the difference between the base supply and the constant load.

With growing 24/7 demand, a fleet of legacy power plants (natural gas, nuclear, coal) is needed to fill in the [significant] gaps left by part-time renewable energy sources. That fleet currently exists, but its overall capacity is declining. Retired plants (to the extent they can be summoned) and new generation will be needed.

However, any new base generation will experience poor capacity factors and difficult gross revenues. Both impair investors' revenues and erode their expected levelized cost of energy. Even if investors overcome profitability concerns, the time it takes to commercialize any new traditional generating asset exceeds the expected demand for new power (extreme example: Georgia Power).

These projections and concerns appear to contradict current trends. Demand has declined in the United States, Europe, and the United Kingdom. Current reporting suggests there could be too much supply, particularly in Europe. However, if projections described by ANS, IEEE, and utilities are correct, the opposite problem could be presented: insufficient supply. If supply becomes the issue as expected, scarcity curves will be taxed, unprofitable generating assets will become profitable, and residential, commercial, and industrial consumers will pay more. This issue is not limited to North America.

Humbert H. writes:

I was listening to an interview of some fund manager from Reno earlier today and he was talking about power shortage around where he lives due to AI server farms. He said they could be quickly and cheaply addressed with new gas powered plants, but due to the Biden administration now requiring all such plants to have complete carbon sequestration this stopped them from being a practical solution.

H. Humbert writes:

Increased the energy supply for data centers is the obvious and near-term brute-force solution. Of course (almost) everybody not in the tech industry assumes that the joule per bit per second for data centers can't be improved and hence producing more energy is the only solution using nuke. In fact Sam Altman said that too, what conventional thinking can possibly go wrong, right?

Zubin Al Genubi asks:

What would be a good way to invest in modern nuclear power? How about Bill Gates project?

Asindu Drileba adds:

I would suspect via buying Uranium ETFs? I first saw this conjecture from following the financier Lyn Alden.

Mark Zuckerberg of recent also mentioned in an interview that Energy and not Compute will be the number 1 bottle neck to AI progress.

H. Humbert responds:

The energy being the presumed AI investment proxy won't last in the long term. Increasing the energy supply is just an incremental engineering no-brainer approach to solve a longer term problem and the approach is not disruptive and it doesn't change the world.

Stefan Jovanovich offers:

Radiant Nuclear
Kaleidos: a Portable Nuclear Microreactor that Replaces Diesel Generators

Peter Penha writes:

A relevant interview on the Hidden Forces podcast with Brian Janous who was hired by Microsoft in 2011 to focus on energy (Google had just hired someone themselves as they thought the cloud might become something) - wound up as VP of Energy.

AI data centers need to be where they can individually draw the electricity of a city like Seattle (800 MWh) - so away from major urban areas - discusses the history of the grid from Sam Insull through to where we are going…also on the efficiency / consumption of AI chips - his view with AI is Jevons Paradox will apply and the more efficient the chips and the (new) grid gets the more consumers will demand.

Jul

7

I found this interesting piece of research:

Ergodicity transformations predict human decision-making under risk

Decision theories commonly assume that risk preferences can be expressed as utility functions, which vary from person to person but are stable over time. A recent model from ergodicity economics reveals that if people want their wealth to grow at the fastest rate they need to adjust their utility functions depending on the dynamics of their wealth. Here, we ask whether humans make such adjustments by exposing them to different wealth dynamics….Together, these results provide evidence that human risk-taking behaviour is sensitive to the dynamical context in which decisions are made and that long-term wealth maximization is an important explanatory principle.

It's about two things concerning the psychology of gamblers. First, is the attitude of risk towards bets that entail absolute returns & absolute losses. An absolute return from a bet is when:
- a gambler is given say $10 if they win
- and if they loose the outcome is -$5
What amount they loose or gain is absolute i.e in dollars terms, $10 or -$5.

Second is a group of gamblers whose risk is phrased in relative terms. A relative return on a bet is when:
- a gambler gains 10% if they win
- and if they loose the outcome is -5%
Rewards & gains are framed as percentages.

The conclusion of the research is that people that gamble in absolute terms take more risk than people that gamble in relative terms. People do change their risk appetite if you present gambles to them differently.

So I have a question for you specs: Which statement would make you more likely to buy Nvidia's stock?
a) $1,000 invested in Nvidia stock at IPO ($0.04) would become about $3,200,000 ( i.e absolute terms)
b) $1,000 invested in Nvidia stock would grow by 327,000% (i.e relative terms)

Personally a) would do a better job at baiting me to buy some stock (the research is true for me). What about you?

Here is a twitter thread by one of the authors with a more simplified explanation.

Jul

6

I just listened to this guy who uses Credit Default Swaps for countries (sovereign CDS) as an indicator to evaluate if a country is good for tourism investments. He claims his methods can make between 15% to 20% annually. The person interviewing (Joe De Sena) was also a trader on wall street for 20 years and asked some questions I liked.

Do all countries have these sovereign credit default swaps? I did some Googling and I could only find a few dozen listed on here. If more countries have them, is there a comprehensive list where I can look them up? I need those of Kenya right now for example.

In this episode, Joe De Sena, chats with his friend Kalojan Georgiev, currently residing in Zanzibar. Kalojan provides an engaging insight into the untapped potential of Zanzibar as a prime investment destination and a wonderful place to live.

Big Al responds:

There is some web data.

Nils Poertner comments:

by and large, CDS on sov not really relevant at all for so many reasons. better to look at traded bonds in USD (or EUR) and look at volume, too. in any case, test everything!

Asindu Drileba adds:

I don't know if anyone here is following what's happening in Kenya but it's falling apart:
- They are heavily in debt (foreign debts are 65% of GDP)
- They government wanted to increase taxes to service the debts
- We are seeing heavy protests in Nairobi & other Kenyans cities
- The new tax Bills have been withdrawn
- Protests are still intensifying
- Interest rates on treasuries are at 19% (I often laugh when I hear Americans complain about 5% interest rates)

I just wanted to know how the Sovereign CDS are pricing the events or if the predicted them.

Nils Poertner writes:

plenty of opportunities coming for EM markets (listed equity) - one needs to do a lot of research, as always (and look beneath the surface and dig deeper and test ideas and express it in a trade and learn and so on). your search, Asindu is your search alone and am not laying out the road map, just saying it is possible.

Jul

3

I found this to be one of the worst books I've ever read; I couldn't even finish it. It felt like a disjointed collection of blog posts, miscellaneous information, and ramblings about catastrophes. However, the topic of portfolio protection through options trading does have its merits. Here are a few observations:

Spitznagel's Track Record: While Universa has shown good results, its success could be attributed to the specific sequence of market events. It raises the question of how it would perform in a more stable, long-term market environment.

Leveraging Equity Risk: The argument that paying a 3% annual "fee" allows for taking on more equity risk is compelling. This is reflected in the fund’s CAGR relative to the S&P 500.

Options Trading During Crises: As someone not deeply versed in options, I'm curious about how traders manage to capitalize on positions at the peak of a crisis without losing the hedge if conditions worsen, thus maintaining their investment mandate.

Relationships with Dealers: Effective trading in size in options seems to require solid relationships with dealers.

Further Reading: I plan to revisit Safe Haven: Investing for Financial Storms to pinpoint other intriguing aspects.

Asindu Drileba responds:

On the contrary, I liked Chaos Kings. Scott Patterson is not a financier so I understand it when you read him expecting him to sound like a financier but he doesn't. He is just a story teller. You can tell this in his earlier book, The Quants. Which was not really about finance but just a story about financiers.

Reading "The Quants" for example reinforced/confirmed my suspicion on the relationship between gambling & financiers. I found it to be a very beautiful story. A beautiful beginning & a beautiful ending.

Quick overview of "The Quants":

- It's starts with this poker tournament organized amongst financiers.
- During the tournament, the author describes the characters traits of the financiers by outlining their attitude towards playing poker
- The book then talks about their character when there are in the market. (mostly when they are winning)
- The financial crash of 2007/08 humbles alot of the cocky characters. Previously humble financiers remained humble (also made money). (By cocky I mean hubris)
- Another poker tournament was held after the 07/08 crisis. And the attitudes financiers had towards each other actually changed.

To me, Chaos Kings is a continuation of The Quants. It has 3 central themes. 1st theme is the human story behind the characters.
- Didier Sornette & his love for motorcycles (whom some people in this thread think is useless)
- Yaneer Bar-Yam getting heart broken by famine in Ethiopia
- Mark Spitznagel's love for goats

2nd theme is about the disconnect between how "non chaos kings" think about markets & "chaos kings" think about markets.
- Mark Spitznagel's philosophy on risk for example is that risk management should not simply be to cap your down side, but to actually increase returns. But predicting crashes is impossible.
- Didier Sornette thinks some huge market disasters can be predicted & tactically mitigated.

3rd theme is seeing how people apply inter disciplinary research to markets.
- Didier Sornette uses techniques used to predict mechanical failure in rocket engines. And applys them to classifying the nature of bubbles in financial markets and when markets are likely to fail.
- Yaneer Bar-Yam has his background in modeling epidemics & pandemics (Ebola, COVID). And he uses the same tool box to predict the likelihood of crashes in the market. One of the tools described in the book is a statistical indicator described as "mimicry".

So, my take away is that Chaos Kings is not really an "investment book". It's just a story about how a certain group of financiers approach market crashes. I found it to be a great source of potential research topics.

Jun

17

Yes, the chart looks like a moonshot. Two things:

1. NVDA is selling for about 37x revs, which looks very expensive. But at that P/S ratio, it's selling for only 70x earnings. The reason is that it's a profit monster, with ttm operating income almost 60% of ttm revs.

2. Maybe tech gurus can comment on this: Looking into GPUs, I find they have a limited lifespan, especially if run 24/7 at high workload, which AI seems to demand. Even ignoring upgrades, I'm thinking the chips NVDA is selling today may need to be replaced in as little as 3-4 years. Maybe I'm wrong, but I don't read about this aspect of their business model.

Humbert H. writes:

What matters is whether anyone will catch up. That is truly an open question. They're the leader, but more so due the inertia of their customers and not because nobody can replicate their technology. That's the thing about super-highly valued tech growth stocks, nobody can predict their situation even two years from now. Starting before the dotcom era, Microsoft has been growing forever, Cisco not as much, Nortel even less so, to put it mildly.

High-end AI chips are replaced due to obsolescence much more so that "wearing out". Some will certainly fail, but less so than the graphics card-type GPUs, and that's not the driving force for the replacement cycle. Trying to decide if Nvidia is properly valued is a pointless exercise. There are always people who will know 100x about the situation, and if they could truly value it properly, they'd find a way to follow up.

H. Humbert comments:

I would think as the computing power of the GPU or TPU (tensor processing unit) increases, the communication bandwidth among the chips, server chassis and server racks is the limiting factor that affects the overall computing speed of the entire AI high performance computing center's performance. The latter is a complicated issue, depending on the data center's server connection topology and so forth. I am sure NVDA knows about the issue and they will come up with a solution to resolve some of the speed bottlenecks either organically or by M&A. As a result, they will come up with next-gen solutions and products undoubtedly. There are designed obsolesces built into the products.

There is another issue. The training of the LLMs requires exorbitant amount of energy that it can't be sustained. The energy trajectory is almost exponential. Somehow these issues need to be mitigated. The increasing amount of energy expended also translates to the huge burden of cooling for the data center. So either NVDA or other companies may come up with the solutions to address some or all of the issues. Long story short, the product lifecycle remains relatively short.

The culmination of these issues and hence the potential solutions are of course good business for those who sell the gadgets. Both of the private sectors and the brain trust of the government and the defense departments worldwide are well aware of these issues and have been working hard to come up with some viable solutions.

Asindu Drileba writes:

There are other areas like Gaming, Molecular Dynamics Simulation, 3D Rendering/Computer Graphics, Video Editing, Crypto Mining that are GPU heavy and expected to grow in the future. As for AI, I think Nvidia riding on the AI hype is a bit precarious. Yes. They have mostly "locked in" AI tooling such that it makes no sense to compete with them.

What makes it precarious, is that a single paper that finally describes how to perform current AI applications with very cheap compute i.e CPU compute. Will destroy a lot the stock. As this problem is more of a software problem and not a hardware problem. So I expect it to move & get adopted very fast if it is actually solved. Several companies like Symbolica are working on such a solution.

Dylan Distasio adds:

No comment on the investing angle or future of the industry, but modern day chips are capable of handling pretty high temperatures for a very long time. Running at 24/7 high work load at a stable temp within the safe zone probably would actually result in a longer life than a gamer situation where the chip is stressed/heated and cooled down repeatedly. In any case, I don't think lifespan and failure due to thermal issues (if maintained properly) is a significant concern with GPUs. They'd be replaced due to obsolescence first.

Humbert Humbert writes:

This is a position for electrical interconnect. The current NVDA Blackwell chiplets are connected with very short electrical interconnects which are reaching their speed limits. The speed resolutions are to bring the optical connects closer to the edges of the chips. A few years ago DAPRA has a program called PIPES is to do just that using optical fibers. I don't recall the spec and I believe it has energy spec in terms of how many femtojoules per bit as it have been recognized a number of years ago that the digital switching energy will become an energy burden. But this solution may eventually run into chip edge real estate problem because of the size of the optical fiber core.

There are limits with the current state of the art digital neural network even though it is the hottest subject in town. Analog neural network may have its niche applications that could compute at higher speed and with lower power consumption. The following is one of the many example programs that the DoD is investing in.

Jun

15

Lots going around about how NVDA dominates; and MSFT, NVDA and AAPL now account for about 20% of the S&P 500. I was curious to see what happened in a toy index and so did an experiment (using R):

1. Create an index of 500 stocks, each with a starting value of $100.

2. Each year, for 40 years, each stock's value is multiplied by 1 + a value randomly drawn from a normal distribution with mean 8% and sd 15%, roughly what you might see with the S&P 500.

3. The starting value of the index was $50,000. The final value after 40 years was $1,152,446.

4. The final summed value of the largest 10 out of 500 stocks was $142,320, or 12.35% of the 500-stock index.

I was curious to see if megacaps would emerge from a simple toy model. I ran it only once, and they did. For me, this is a comment on the perennial alarm stories about "Only X% of stocks account for Y% of the market!" Even with a simple model, you wind up with something like that.

Adam Grimes agrees:

Can confirm. Have done variations of this test with more sophisticated rules, distribution assumptions, index rebalancing, etc. Get similar results.

Peter Ringel responds:

so we can take this ~12% of the index as a base value, that develops naturally or by chance? Then a clustering of being 20% of a total index (only greater by 8%) does not look so outrageous.

William Huggins is more concerned:

keep in mind it's 10 companies making up 12% (~1.2% each) vs 3 companies making up 20% (8.3% each) - in that sense, the concentration DOES look pretty high. am reminded of when NT was 1/5 of the entire CDN index in 99/00.

Peter Ringel replies:

You are right, I failed to catch this difference of only 3 stocks. In general, I am not so much surprised about the concentration. Money always clusters. Always clusters into the perceived winners of the day. Should they blow up, money flows into the next winner. To me, the base for this is herd mentality.

Adam Grimes comments:

It's Pareto principle at work imo. I'm not making any claims about exact numbers or percents, but as you use more realistic distribution assumptions (e.g., mixture of normals) the clustering becomes more severe. There's nothing in the real data that is a radical departure from what you can tease out of some random walk examples. Winners keep on winning. Wealth concentrates. (As Peter correctly points out.)

Asindu Drileba offers:

Maybe you try replacing the normal distribution of multiples with a distribution of multiples constructed with those historically present in the S&P 500? It may reflect the extreme dominance in the market today.

To me, the base for this is herd mentality.

It is also referred to as preferential attachment:

A preferential attachment process is any of a class of processes in which some quantity, typically some form of wealth or credit, is distributed among a number of individuals or objects according to how much they already have, so that those who are already wealthy receive more than those who are not. "Preferential attachment" is only the most recent of many names that have been given to such processes. They are also referred to under the names Yule process, cumulative advantage, the rich get richer, and the Matthew effect. They are also related to Gibrat's law. The principal reason for scientific interest in preferential attachment is that it can, under suitable circumstances, generate power law distributions.

Zubin Al Genubi writes:

Compounding of winners is also at work and returns will geometrically outdistance other stocks. No magic, just martini glass math.

Anna Korenina asks:

So what are the practical implications of this? Buy or sell them? Anybody in the list still owns nvda here? If you don’t sell it now, when?

Zubin Al Genubi replies:

Agree about indexing. Hold the winners, like Buffet, Amazon, Microsoft, NVDIA. Or hold the index. Compounding takes time. Holding avoids cap gains tax which really drags compounding. (per Rocky) Do I? No, but should. It also works on geometric returns. Avoid big losses.

Humbert H. wonders:

But what about the Nifty Fifty?

Jun

4

Top-Level Domains | The Economics of Everyday Things

Those letters at the end of web addresses can mean big bucks — and, for some small countries, a substantial part of the national budget. Zachary Crockett follows the links.

Asindu Drileba writes:

Mali owns the .ml domain for that may do well for Machine Learning. I wonder why they are sleeping on it?

Jun

1

Numberphile Video demonstrating that a cone that is 80% full in height is actually 50% full in volume. You will also know if your getting scammed in a bar.

Cones are MESSED UP - Numberphile

Zubin Al Genubi writes:

This is why convexity, compounding, and geometric or exponential growth are hard to comprehend.

Kim Zussman comments:

Geometric returns are important when assessing performance. From an investor's perspective, average returns underweight when a manager loses everything (because it is sum-based), but geometric returns don't (because it is a product).

Geometric mean

May

22

Learning skillful medium-range global weather forecasting, from Google DeepMind:

Global medium-range weather forecasting is critical to decision-making across many social and economic domains. Traditional numerical weather prediction uses increased compute resources to improve forecast accuracy but does not directly use historical weather data to improve the underlying model. Here, we introduce GraphCast, a machine learning–based method trained directly from reanalysis data. It predicts hundreds of weather variables for the next 10 days at 0.25° resolution globally in under 1 minute. GraphCast significantly outperforms the most accurate operational deterministic systems on 90% of 1380 verification targets, and its forecasts support better severe event prediction, including tropical cyclone tracking, atmospheric rivers, and extreme temperatures. GraphCast is a key advance in accurate and efficient weather forecasting and helps realize the promise of machine learning for modeling complex dynamical systems.

Asindu Drileba writes:

I looked and how they split the surface of the earth into tiny meshes in order to form a Graph upon which the rest of the model is built. This technique's description looked familiar to another technique called "Numerical Weather Prediction" or "NWP". The paper by Deep Mind does have several papers referencing "Numerical Weather Prediction" or "NWP".

I learnt about NWP from this PBS Nova documentary, Prediction By The Numbers. The documentary has many descriptions on tools for predicting. Wisdom of the Crowds, Probability Theory, and Numerical Weather Prediction. Meteorologists were interviewed and they showed their NWP programs in action (it is shown how the model generalized with time to predict a more accurate forecast of the in the weather). The meteorologist says it is their best model and also goes ahead to say that "it works so well."

May

18

Lots More on How CHIPS Act Money Got Awarded

In 2022, Congress passed the CHIPS Act, which set aside tens of billions of dollars in loans and grants in order to encourage companies to build new semiconductor fabs in the United States. We're still very early in the process. It's going to be a long time before we know if the US will become a major player again in the production of advanced chips. But the process is well underway and the bulk of the awards have been officially announced, with much of the money going to Intel, Samsung, TSMC, and others. So how did the grants get allocated — and what's next?

While it is clear that money earning or losing events like quarterly earnings announcements have an impact on the market (stock prices). I am not sure if government subsidies & grants have an impact on stock prices. Is there a tool that can be used to track events related to government subsidies & grants?

Big Al responds:

An interesting question. Probably start here:

USASpending.gov

USAspending is the official open data source of federal spending information, including information about federal awards such as contracts, grants, and loans.

Also interesting research tracking stock trades by members of the US Congress:

Capitol Trades
Smart Insider
Senate Stockwatcher

May

4

The Doctor Is In. And He’s an Orangutan.

For the first time, researchers have seen a wild animal treat an open wound with a medicinal plant. After getting injured—probably in a brawl with another male—a wild Sumatran orangutan chewed the stems and leaves of a vine humans use to treat wounds and ailments such as dysentery, diabetes and malaria. The orangutan then repeatedly smeared the makeshift salve on an open gash on its cheek until it was fully covered. After the treatment, scientists saw no signs of infection. The wound closed within five days. And it healed within a month.

Jeffrey Hirsch is enthusiastic:

This is awesome! An good friend of mine spent several years in Borneo working with Orangutans under Birute Galdikas’ program. They are super crafty and smart. Don’t doubt this.

Humbert H. writes:

And nobody can explain how they know to do this in these situations. There is obviously a lot of learning apes can acquire from others, but this? There is also no way the current understanding of how genetic information is passed on that can explain this. There is something very mysterious about the mind and animals doing non-obvious things is the best example, this is not a simple biological phenomenon.

Asindu Drileba comments:

One of the things I hear in the AI research community in the pursuit of of AGI (Artificial General Intelligence) is people thinking of intelligence as something hierarchical like height.

In The Singularity is Near Raymond Kurzweil makes a plot of Computers approaching AGI. He puts insects at the bottom and manuals later then humans at the top. You often hear some people say that "We haven't yet reached dog level AI, so we can't say we can reach human level AI soon." That statement makes the assumption that A humans intelligence is more than that of a dog. But it has been reported in some cases a dog's sense of smell can be 100,000 more acute than that of a human being! And not just that it can tell time just by smelling what's around. Another example is also how birds can sense magnetic fields and use them like a compass.

Anyway my point is that just by the (limited) way humans perceive reality we have access to some secrets we can't pass to animals. My suspicion is that animals also have their own secrets that they cannot pass to us.

Humbert H. adds:

They have recently discovered that some insects are self-aware. The test that's used for animals is that they recognize their reflection in the mirror as themselves judging by their reaction. Usually only dolphins, apes, and some corvids (crows) pass the test.

But more importantly, what I meant was that animals seem to "know" how to do things that no current scientific understanding can explain. This means we don't understand basic things about animal (and human) mind. AI is a machine function: an algorithm using some data provides some outputs in response to inputs. A mind is like that too, except we really don't understand the nature of self-awareness, nor do we understand how animals just "know" things. Sometimes they call it "instinct" but there is no real science behind that word. And in this case it's not even that, apes have no "instinct" to cure wounds with specific processed plant material.

Jeff Watson writes:

Here is an interview with cognitive psychologist, Donald Hoffman. Some find him brilliant, some a flake. His ideas are unconventional to say the least, but the questions that come to mind out of his interview will break one’s brain. Many moments in the video, I pause and ask myself how this applies to markets.

Stefan Jovanovich gets philosophical:

The wheel of time turns on the axle of our self-awareness: Transcendentalism.

May

2

Smörgåsbord for the beginning of May

Zubin Al Genubi recommends:

Market Tremors: Quantifying Structural Risks in Modern Financial Markets

Clear exploration of potential causes of and prediction of volatility events caused by Dominant Agents. Explores imbalances created by ETFs ETNs Banks, FED Market Makers.

Asindu Drileba suggests chaos:

Doyne Farmer describes the relationship between Roulette Wheels, The Weather, Financial Markets, and Economies as a whole. He thinks companies that don't make the energy transition from fossil fuels will all go bankrupt in the next 5 years. He is also promoting his new book:

Making Sense of Chaos

Here is the discussion:

Simplifying Complexity: Making sense of chaos with Doyne Farmer

Nils Poertner points to probability:

Birthday problem

stochastics is really quite counter-intuitive - it deals with "uncertainty" rather than basic algebra or geometry which one learns in schools. good training ground for learning about markets as well. (always found that stochastics often attracts folks who are a bit off the normal conventions / and have an genuine curiosity in things rather than go with what is fashionable)

Apr

27

From Asindu Drileba:

This video is about how to use a technique known as "dimensional analysis" that can be used to derive equations or attain further insights about a physical system.

From Big Al:

A nice, simple explanation of Markov chain.

(Sidebar history note:Alexandre-Théophile Vandermonde.)

More complex: Markov Decision Processes - Computerphile.

Apr

23

Looks like someone in Can Gov was listening in for ideas (Tax rate to rise from 50% of reg to 67% of reg):

Capital gains tax change draws ire from some Canadian entrepreneurs worried it will worsen brain drain

In the 2024 budget unveiled Tuesday, Finance Minister Chrystia Freeland said the government would increase the inclusion rate of the capital gains tax from 50 per cent to 67 per cent for businesses and trusts, generating an estimated $19 billion in new revenue. Capital gains are the profits that individuals or businesses make from selling an asset — like a stock or a second home. Individuals are subject to the new changes on any profits over $250,000.

Big Al is sanguine:

No worries - it only affects a few:

The government estimates that the changes would impact 40,000 individuals (or 0.13 per cent of Canadians in any given year)…

H. Humbert writes:

With 67%, the government clearly thinks that either it both needs and deserves the profits of some people more than they do OR that those people need to be treated like one would treat an enemy, without any regard for their needs or feelings. Let's see, would a Communist think that way (both ideas) about his or her class enemy?

William Huggins explains:

It's a move back towards the status quo ante 1980s tax cut. The idea that tax cuts are only good is just silly. As silly as the notion that government is efficient with those same taxes. This isn't revolutionary, simply the slow reduction of a subsidy we -thought- would lead to more investment. Turns out future demand is a larger determinant of that than current taxes. We gave too much to capital back in the early 80s when we rebalance last time and now were rebalancing again. Cap gains will still pay less tax than working folks. No need for enemies or "communists".

H. Humbert replies:

I apologize William, the problem was my reading comprehension as I wasn't familiar with the meaning of the term "inclusion rate" in the Canadian tax system and interpreted it incorrectly after, to be honest, spending about 20 seconds to "read" the article. With your explanation and the tiniest bit of research, this makes sense. As I mentioned before, I'm against special cap gains rates, but only if (a) the losses aren't capped (b) there is no special "investment gains" tax as currently exists in the US.

Asindu Drileba adds:

David Graeber once mentioned that the most productive period in American industry was when the tax rates were highest (65%). The referenced the advances made by Bell Labs as a example. He claimed that the productivity occurred because corporations were nudged by the high taxes to invest more money into research and development.

H. Humbert provides context:

Very few people paid the top marginal rate as tax shelters were highly prevalent and a lot easier to use than they are now.

Hernan Avella comments:

True MMT’rs would argue that rates should be 0 and the tax rate higher, as needed to curb inflation.

Apr

16

Investors wrongfooted as ‘higher for longer’ rates return to haunt markets

Zubin Al Genubi asks:

Interest alone on US debt is 1 trillion dollars a year! Anyone concerned?

Larry Williams is definite:

NOPE. NOT AT ALL.

Art Cooper, however:

*I* am certainly concerned, in the long term. When the coverage ratio on gov't debt auctions drops close to 1.0, it will be time to take meaningful action, with a major re-allocation of investment portfolios.

Larry Williams responds:

Not to worry…says MMT guys…as long as we are not gold-backed $, it's all just accounting numbers.

Kim Zussman wonders:

Reallocate to what? (he says looking around twice with stocks near ATHs)

Art Cooper suggests:

There are a universe of hard assets out there, including gold (though GLD could easily go far higher). Because I like to emulate the Sage and shop in the bargain basement, I personally find extremely distressed income-producing real estate of interest. Babies are being thrown out with the bath water.

Larry Williams writes:

The public debt is just $ in savings accounts at the Federal Reserve Bank. When it matures the Fed transfers those dollars to checking accounts (aka reserves) at the same Fed. It's just a debit of securities accounts and a credit of reserve accounts. All internal at the Fed. When gov sells new Tsy secs, the Fed debits the reserve accounts and credits securities accounts. Those $ only exist as balances in one account or the other.

Asindu Drileba adds:

David Graeber once mentioned that the US can never default on its debts since the Fed is the largest holder of Treasuries.

William Huggins comments:

its not that the US -can't- default on its debts, its that 70% of those debts are to americans. so what is the probability of americans voting to default on themselves when they have the ready alternative of printing money? more important might be whether or not the 30% foreign holders will keep playing along but that analysis is an exercise in ranking "next best alternative" for them. when one starts looking under the hood at the alternatives, its boils out like china's bank regulator said in early 2009, "except for treasuries, what can you hold? gold? you don't hold japanese government bonds or uk bonds. us treasuries are the safe-haven. for everyone, including china, it is the only option: "we hate you guys but there is nothing much we can do."

H. Humbert replies:

The Americans would be about equally unlikely to default if most of the debt was held by foreigners. If you can print money there is no need to piss off any of your "customers". It's not like things worked out super well for Argentina, at least until they hit bottom.

Apr

15

From Easan Katir:

The Hall of Uselessness: Collected Essays, by Simon Leys.

Simon Leys is a Renaissance man for the era of globalization. A distinguished scholar of classical Chinese art and literature and one of the first Westerners to recognize the appalling toll of Mao’s Cultural Revolution, Leys also writes with unfailing intelligence, seriousness, and bite about European art, literature, history, and politics and is an unflinching observer of the way we live now.

From Zubin Al Genubi:

Pathogenesis: History of the World in Eight Plagues, by Jonathan Kennedy.

According to the accepted narrative of progress, humans have thrived thanks to their brains and brawn, collectively bending the arc of history. But in this revelatory book, Professor Jonathan Kennedy argues that the myth of human exceptionalism overstates the role that we play in social and political change. Instead, it is the humble microbe that wins wars and topples empires.

From Asindu Drileba:

Math Without Numbers, by Milo Beckman.

Math Without Numbers is a vivid, conversational, and wholly original guide to the three main branches of abstract math—topology, analysis, and algebra—which turn out to be surprisingly easy to grasp. This book upends the conventional approach to math, inviting you to think creatively about shape and dimension, the infinite and infinitesimal, symmetries, proofs, and how these concepts all fit together. What awaits readers is a freewheeling tour of the inimitable joys and unsolved mysteries of this curiously powerful subject.

Peter Ringel is watching:

Voltaire: The Rascal Philosopher

I discovered a terrible knowledge gap and missed details of a great one. so many angles to be impressed. his writings seem to be the least of it. he even gamed the king's lottery and won with a group of investors & mathematicians.

William Huggins suggests a somewhat older work:

A General History of The Most Prominent Banks, by Thomas H. Goddard, published in 1831.

its dry - but if you are interested in the 1819 panic, there are some good details. the book is mistitled imo as 3/4 of its pages and 2/3 of its text centers on the history of central/national banking in the united states from 1786 through 1831 (publication). on titular matters, it had a couple of interesting tidbits on the bank of genoa and some "interesting" statistical information for archivists but there are better modern sources on major banks in venice, the netherlands, england, and france (for example, the author skips over how the bankers of geneva funded the french revolution to knock the bank of genoa off its perch, etc). i suspect such deficiencies are because the text was designed as ammo in the "bank wars" of the early 1830s rather than a deep exposition on titular topics.

its exposition on us matters feels remarkably haphazard, i presume because the author's intended audience would have the context to appreciate why it includes what it does, including a description of the bank of north america, hamilton's report to congress on the need for a bank, and a brief on the First Bank of the US. where it begins to shine is in the next set of docs, which includes an auditor's report and statement by the president of the Second Bank of the US on how the panic of 1819 was navigated. it follows with mcduffie's 1930 report to congress on the SBUS (includes more details on the rise and fall of FBUS), and closes with a statistical archive of the "monied institutions of the US" and an appendix on how banking and commercial exchange granularly worked in the 1800s.

Stefan Jovanovich comments:

I was puzzled by the "decline and fall" description, since the Bank did not fail but simply had its charter expire without renewal because George Clinton did not like what Thomas Willing had done as President of the Bank. (Clinton failed to cast what would have been the winning vote for renewal.)

William Huggins responds:

"fall" referring to its near brush with survival, not any sort of mismanagement or fraud as in 1819. mcduffie describes FBUS as the victim of partisan politics, but one of such import that the same party who killed it started calling for a replacement almost immediately.

Stefan Jovanovich adds:

They wanted what Willing would not give them - a central bank that would do what the Fed does now - discount the Treasury's IOUs at par. Can't have a war without that.

Apr

13

This is a topic that keeps appearing when people talk about probability. I don't seem to have a good intuition for it. Is the stock market with memory or without memory? Why? What would be your intuitive explanation of what memory is?

From Memorylessness:

In probability and statistics, memorylessness is a property of certain probability distributions. It usually refers to the cases when the distribution of a "waiting time" until a certain event occurs does not depend on how much time has elapsed already. To model memoryless situations accurately, we must constantly 'forget' which state the system is in: the probabilities would not be influenced by the history of the process.

Only two kinds of distributions are memoryless: geometric distributions of non-negative integers and the exponential distributions of non-negative real numbers.

Humbert H. responds:

Of course it's not completely memoryless otherwise there would be no point to any spec of this list trying to beat the market. It's ALMOST memoryless, and that's why it's hard to beat, but there are still some irregularities, like days of the week, month, season, reaction to events, like increased volatility following a big change. It would have a lot more memory if people didn't try to take advantage of the irregularities, because market participants have emotions and also information doesn't spread instantaneously even in this day and age.

Eric Lindell comments:

Blackjack is with memory, provided the number of decks is finite. As you play with more and more decks, the game becomes less memory-dependent. A small player in a huge market makes trades that are less memory-dependent than a big player's trades. The bigger the portion of the total market a trader trades, the more memory-dependent it becomes.

Wikipedia's discussion of a memoryless probability distribution refers to a poisson process. The time before the next car arrives at a toll booth doesn't depend on the time since the last car arrived — provided the cars' arrivals are truly random. This would NOT be the case with a nonrandom distribution, as when more cars arrive per minute during rush hour.

Zubin Al Genubi writes:

A normal distribution of a series of events, indicates that the events are independent of each other, in that the occurrence of one does not affect the probability of another. Of course the market has memory and emotion. We are looking for the regularities to trade that are not random with a high degree of confidence.

Larry Williams agrees:

Amen! People react in similar fashion to events and those reactions create patterns. Plus, there are unique time elements to many markets; jewelry is mostly sold at Christmas, hogs live and die in 18 months etc.

Penny Brown adds:

Investors who suffer a big, sudden decline in a stock remember it. Often they vow to hold on until they are made "whole". This can cause a stock to sell off as it approaches that spot. But if the stock clears this area, the weak hands are gone, and the stock can move up sharply.

Big Al suggests:

For further study, re the quality of "memoryless" and possible applications:

Markov chain

Hidden Markov model

Also, Vic has referred to Markov processes relating to the market calendar at the top of this site.

Apr

7

Kim Zussman is optimistic:

Age and High-Growth Entrepreneurship
Pierre Azoulay, Benjamin F. Jones, J. Daniel Kim, and Javier Miranda
American Economic Review: Insights
Vol. 2, No. 1, March 2020

Abstract

Many observers, and many investors, believe that young people are especially likely to produce the most successful new firms. Integrating administrative data on firms, workers, and owners, we study start-ups systematically in the United States and find that successful entrepreneurs are middle-aged, not young. The mean age at founding for the 1-in-1,000 fastest growing new ventures is 45.0. The findings are similar when considering high-technology sectors, entrepreneurial hubs, and successful firm exits. Prior experience in the specific industry predicts much greater rates of entrepreneurial success. These findings strongly reject common hypotheses that emphasize youth as a key trait of successful entrepreneurs.

Asindu Drileba adds:

This infographic is really good.

Big Al finds value in experience:

Neoclassical Theory Versus Prospect Theory: Evidence from the Marketplace
John A. List
ISSUE DATE June 2003

Abstract

Neoclassical theory postulates that preferences between two goods are independent of the consumer's current entitlements. Several experimental studies have recently provided strong evidence that this basic independence assumption, which is used in most theoretical and applied economic models to assess the operation of markets, is rarely appropriate. These results, which clearly contradict closely held economic doctrines, have led some influential commentators to call for an entirely new economic paradigm to displace conventional neoclassical theory e.g., prospect theory, which invokes psychological effects. This paper pits neoclassical theory against prospect theory by investigating three clean tests of the competing hypotheses. In all three cases, the data, which are drawn from nearly 500 subjects actively participating in a well-functioning marketplace, suggest that prospect theory adequately organizes behavior among inexperienced consumers, whereas consumers with intense market experience behave largely in accordance with neoclassical predictions. The pattern of results indicates that learning primarily occurs on the sell side of the market: agents with intense market experience are more willing to part with their entitlements than lesser-experienced agents.

Mar

24

An alternate understanding of a market being at all time high (market reaching new prices it has never encountered) is this: "Everyone that has ever bought that stock or instrument is now in profit". What might be the psychological implications of this?

Kim Zussman comments:

It is possible (and probable) to buy, then sell after a decline and stay out only to see it reverse and go up further. This (timing) is one reason it is so much easier to do better with B/H than trading.

Big Al adds:

The other advantage to B&H is that the opportunity cost viz time/attention required is basically zero. I have looked at various index timing approaches and have not found anything that beats B&H, especially when considering the vig and opportunity cost. However, should one need to scratch the itch, timing strategies may work better with individual stocks. But again, opportunity cost.

Humbert H. writes:

I've always been believer in B&H vs. trading. But even in B&H the debate between indexing and individual stock selection never dies. I don't like indexing, but I don't have a mathematical basis for that. It's a fundamental belief that buying things without any regard to their economic value has to fail in time, at least relative to paying some attention to it.

Zubin Al Genubi adds more:

Another aspect of buy and hold that Rocky pointed out is the capital gain tax severely eats into returns. The richest guys hold for years and have only unrealized untaxable gains.

Art Cooper agrees:

There was an excellent article in the Jan 7, 2017 issue of Barron's by Leslie P. Norton on VERY long-lived closed end mutual funds which have surpassed the S&P's performance. They have all followed buy and hold strategies.

Michael Brush offers:

Far more money has been lost by investors in preparing for corrections, or anticipating corrections, than has been lost in the corrections themselves.
- Peter Lynch

Steve Ellison brings up an important point:

And yet trading is one of the focal points of this list. The way I square this circle is to keep most of my trading account in an equity index fund at all times. When I think I have an edge, I make trades using margin.

Larry Williams writes:

B&H is the keys to the kingdom, but…the massive fortunes of Livermore were short term trades despite his comment about sitting on your hands. Even the current high performers, Cohen, Dalio, Tudor etc use market timing. When I won world cup trading $10,000 to $1,100,000, it was all about timing and wild crazy money management. One approach wins big the other wins fast. A point to ponder.

Bill Rafter writes:

What we found in studying only the SPX/SPY is that in the long run a buy-and-hold yielded 9.5 percent compounded annually. That was from 1972 to recent. Our argument is that studies before 1972 are flawed. That 9.5 was great considering there were several collapses of ~50 percent. However if you could just eliminate the collapses you could raise the return to 13.5 percent compounded annually.

Eliminating the down moves did not involve prescience. You did not need to forecast recessions, only identify them when you were in one. That was not difficult, and timing was not a critical as one might think. We identified several algos that worked well.

When you were out of equities, you could either simply hold cash, or go long the 10-year ETF. The bonds were better, but not by much. Interestingly, long term holding of bond ETFs yielded low single-digit returns. Best avoided. Which also means that the Markowitz 60/40 strategy was a sub-performer.

Taxes are investor/vehicle specific. For example, if you use a no-tax vehicle, there are no taxes. Regarding turnover, there are very few transactions, as there are very few recessions. The strategy is basically B&H, but with holidays.

Asindu Drileba has concerns:

My problem with buy & hold Is that it has no risk management strategy. If you bought the S&P 500 in 1929 for example during the wrong month. It took you 25 years i.e until 1954 not even to make profit, but just to break even. The real question is, how do you know your not investing in a market path that will take 25 years just to break even?

Humbert H. responds:

That’s why, dollar cost averaging. I don’t think anyone thinks buy once in your lifetime and never interact with the stock market ever again. I think if you had averaged in monthly or quarterly from the summer 1929 through summer 1959 and then held and lived off dividends or cashed out/interest in retirement, you did well.

Art Cooper adds:

The year 1954 is almost universally given as the "break-even" year to recoup losses for buy & hold investors who bought at the 1929 peak. It's wrong to do so. First, it ignores dividends. Had dividends been re-invested the recovery year would have been much earlier. Second, it ignores the deflation which occurred during the Great Depression. In this column Mark Hulbert argues that someone who invested a lump sum at the 1929 peak would have recovered in real economic terms by late 1936.

I'm not arguing against dollar-cost averaging, merely pointing out a historical falsehood.

Hernan Avella writes:

What people should do while they are young and have human capital left is to leverage!

Life-Cycle Investing and Leverage: Buying Stock on Margin Can Reduce Retirement Risk

The most robust research, incorporating lifecycle patterns and relevant time horizons for long term investors tells us that the optimal allocation is 50/50 all equities, domestic and international. But most ppl don’t have the gumption to be 100% on equities.

Mar

13

I found this podcast episode very interesting.

Contextualization Within a Framework of Conditional Probabilities w/ Will Gogolak

As a risk officer with the Chicago Mercantile Exchange, Will Gogolak was setting margin requirements and saw a wide variety of traders’ accounts and what separated the winning traders from the losing ones, before leaving to pursue his own trading and obtaining a PHD in finance and share his knowledge of quantitative analysis and market experience with students at Carnegie Mellon University. Combining his market experience with knowledge of statistics helps William create his custom buy the dip strategy with futures and leveraged ETFs, and focusing on probabilities and determining market direction for informed trading decisions.

Peter Ringel agrees:

Love the hole series. Half of speclist was a guest.

Big Al offers:

Also very informative are these interviews with "Uncle Roy", on Top Traders Unplugged:

From 2014:
Part 1
Part 2

From 2023

Zubin Al Genubi comments:

Speaking of cognitive bias, I realize that if I feel bearish, so does everyone else. You have to go against how you feel and against the consensus.

Sam Johnson asks:

Do you need to go against the cognitive bias of how everyone FEELS or how everyone is positioning?

Zubin Al Genubi responds:

Don't most traders and their systems trade and position for that past regime? As Roy said, trend followers are all piled in at the turns and all will reverse at the same time. With the widespread use of systems everyone is doing basically the same trade. You can't get a fill after the turn as we saw last fall. You have to pre-position…be in position ahead of the enemy forces.

Feb

29

Nvidia Hits $2 Trillion Valuation on Insatiable AI Chip Demand

The chips are so valuable that they are delivered to the networking company Cisco Systems by armored car, said Fletcher Previn, Cisco’s chief information officer, at The Wall Street Journal’s CIO Network Summit this month.

H. Humbert is skeptical:

This won't end well, but I have no idea about the timing. I have a mixed record on predicting the future, so my prediction is worth what you paid for it, but this is what's likely to happen: due to the chip shortage (the TSMC bottlenecks described aren't easily solved in the short term) and their high prices, NVIDIA's hold on the software stack will be punctured. Someone will say "Hey, we need a second source, it's not good to just have one supplier". Once that happens their monopoly will be over, and it will deflate. Are there any signs of this today? No, none.

Asindu Drileba writes:

Nvidia's edge will evaporate if there is a breakthrough in a new AI paradigm that is not as computationally intensive as deep learning. Herding exists in research just as it does in markets. As of today, researchers are herding on deep learning because it is what has shown a great track record so far. But it is clearly known that there are better (but unarticulated) ways to build systems that exhibit the properties of Artificial Intelligence that industry wants to use to solve problems. As long as these techniques are not yet developed. I still see a growing market for someone like Nvidia in the long term.

H. Humbert adds:

Nvidia will see a growing market for a long time to come, the point is they're not levitating due to durable hardware advantages but because nobody wants to abandon their CUDA toolkit. Not yet, but some day someone will diversify for any number of reasons. They will still remain king of the hill, but cracks will develop.

Humbert H. comments:

Von Neumann latency and huge power consumption are issues and will eventually be a big enough problem. It is a know problem. If not solving the problem organically, I am sure they are looking out to buy the solutions if there are viable solutions. Don't know when it happens but will happen.

Jensen Huang's speech in 2011 about failure and changing course quickly. Sounds like a trader mindset.

Some alternate techniques are being developed but most of the average Joes don't know that yet. Speaking from my observation of what are happening, not just sheer speculations. This conference ISSCC - International Solid-State Circuits Conference on solid state device held this week at SF definitely covered areas related to high speed solid state device advances, limitations and solutions. The published papers and abstracts should have the most updated information.

Yelena Sennett is skeptical, too:

As long as Nvidia are buying their own chips, their sales will keep growing, especially if they keep recording it as revenue before delivery, lol. Scott McNealy's famous 'What were you thinking?' rant to investors for bidding Sun Microsystems' stock price up to 10x sales during the DotCom bubble:

At 10 times revenues, to give you a 10-year payback, I have to pay you 100% of revenues for 10 straight years in dividends. That assumes I can get that by my shareholders. That assumes I have zero cost of goods sold, which is very hard for a computer company. That assumes zero expenses, which is really hard with 39,000 employees. That assumes I pay no taxes, which is very hard. And that assumes you pay no taxes on your dividends, which is kind of illegal. And that assumes with zero R&D for the next 10 years, I can maintain the current revenue run rate. Now, having done that, would any of you like to buy my stock at $64? Do you realize how ridiculous those basic assumptions are? You don't need any transparency. You don't need any footnotes. What were you thinking?

It’s not different this time - trading around ~ 30 times sales! The only question is if the market is different this time and NVDA is just one stock that will not affect the general market when it goes down back to reality of $200 - $300.

Feb

28

Fractal scaling and the aesthetics of trees

Trees in works of art have stirred emotions in viewers for millennia. Leonardo da Vinci described geometric proportions in trees to provide both guidelines for painting and insights into tree form and function. Da Vinci’s Rule of trees further implies fractal branching with a particular scaling exponent.

H. Humbert writes:

I could never understand how fractals help with markets. Yes, the world is fractal, but fractals are essentially a way to describe the "roughness" of random patterns. But is this roughness permanent? No. Are the patterns predictable? No. Yet somehow some wiggles are described as bullish and bearish fractals. Sounds like snake oil to me.

Asindu Drileba responds:

You're right! Mandelbrot himself admits that his techniques cannot predict the direction a financial instrument will move. He however says that his techniques can predict "by how much" a financial instrument will move. He describes that "large movements are more likely to be followed by large movements" and "small movements are more likely to be followed by small movements." Here is a short video of Mandelbrot describing his model.

H. Humbert replies:

Never read his books. I know Victor hated him with passion, he was one of the three most guilty, the other two were Taleb and Buffett. Watched the video, a lot of words but nothing practical. Also since his mode of thinking is simple and algorithmic, and he is famous, if there ever was anything to be gained from it, by now algorithmic trading surely made all those possible gains disappear.

Laurence Glazier comments:

There is always an element of hand-waving in attempts to make things easier than they are, and it can be seductive. Nature, however, likes economy of means, and therefore if the same-ish pattern can be used at different scales, I would expect this to happen - but this assertion itself has an element of hand-waving.

H. Humbert adds:

To me the main element of hand-waving is that coastal topography and tree branch patterns created by very different mechanisms themselves have anything to do with predicting market moves where human psychology among many things is involved.

Zubin Al Genubi writes:

There are entire financial industries and degrees relating to prediction, measurement, and trading volatility. It is one of the most important aspects of trading and protecting yourself from ruin. A simple example of the importance of understanding volatility is its mean reversion. In time of stress and price drops this is a key.

Anatomy of a Meltdown: The Risk Neutral Density for the S&P 500 in the Fall of 2008, Justin Birru and Stephen Figlewski.

September 2008 was when the crisis hit in force….On 55% of the trading days in October and November 2008, the index moved more than 3% up or down (corresponding to annualized volatility in excess of 47%). Interestingly, while it is well-known that the market tends to move faster and further on the downside, in this extraordinary period sharp moves to the upside were just as common. On the two days with the largest price changes in October, the market rose more than 10%.

H. Humbert continues:

Once again something unpredictable happened that was difficult to take advantage of. It seems like crisis-related volatility would have to subside sooner or later when the crisis is over, is this a revelation? I recall the March 2020 day when the market hit the Covid lows. I literally said to myself "this has got to be the bottom". But did I do anything? No, because I really wasn't sure. Some forces ended the crisis, but they're only obvious in retrospect.

Feb

7

It is not hard to see this is very late stages of speculative madness but I really would like to know how the risk management teams approve buying Nvidia stock here after adding $200 billion to market cap in 3 days?

Larry Williams offers:

Maybe my cycle forecast for NVDA would help:

Asindu Drileba writes:

I don't know why people are still buying Nvidia. But this is what I personally think of the stock. Nvidia has an 80% market share in the Graphics Card business. Their bread and butter used to be video gaming, 3d animation, video editing, later crypto mining, AI (computer vision), AI (Large Language Models), AI (Image generation) possible new advances may occur in Molecular Dynamics, Self driving cars etc. The CEO had an interesting interview where he talked about possible areas Nvidia may venture into.

But here is one strange thing about high performance computing (Nvidia's Niche): We would think that the better (higher performing) their products are, the less people would buy because people would do more with less right? It's actually the opposite.

— In gaming for example, when graphics cards improved people moved to less polygon looking characters and wanted more details like finer hair & plants. From there they even went to more computationally intensive algorithms like ray tracing that mimic real world scattering of light. Requiring even more compute in subsequent algorithmic advances.

— In Bitcoin, many people using Nvidia GPUs made it more difficult to earn money from crypto mining. Which requires people to have even more Nvidia GPUs just to continue earning the same income.

— In AI, when ever a new breakthrough was made, researchers often trained models with larger datasets, using more & more GPUs. Chat GPT for example was trained on 1 Trillion corpus of text.

So if they do maintain this 80% market share and these underlying industries continue to grow (and make new break throughs). It makes sense that Nvidia will be very valuable in the near or distant future. Buying now (at all time highs) is definitely dangerous but, even if the bubble pops, the underlying industries it facilitates will still be present. And if more breakthroughs in these industries are made, it makes sense that Nvidia still has some value left in it.

Cagdas Tuna responds:

Good fundamental points and there I have 2 counter outlook:

-Gaming industry; I almost everyday play an online game called Destiny 2, and their developer Bungie has reduced workforce around 10%. I know many other gaming companies are reducing/reduced workforce which doesn't give too much optimism in that area.

-Bitcoin mining; there is halving in a few weeks and this will require more powerful computers but it will also increase the cost which in the end will end up new miners losing money in most cases. Only way to maintain gains in mining is Bitcoin price to double or triple in a year.

Even on the best possible scenario it will not add 200 billion dollars worth growth in many many years.

Steve Ellison comments:

Words of wisdom from Rocky's Ghost, posted in the Spec List on April 4, 2017. And yes, I am long NVDA. I believe this is the study Rocky referred to.

Soros and I share very little. However, I have come to agree with him that the right position is to be long "bubble" (however defined). I used to subscribe to Anatoly's view and to be bearish during bubbled but I discovered that from a risk-adjusted-return perspective, it's better to be right "today" than right "tomorrow." Along this point, I read a study that shows a substantial percentage of stock returns occur during the last surge in a "bull market". If you miss this surge, it's very difficult to keep up with the indices in the short term. And in the long term, we're all dead.

Asindu Drileba replies:

Gaming Revenue was about $142B just in 2022. If cloud gaming, something Nvidia is planning todo is successful, I expect this to jump by several multipliers. I expect Cloud gaming to be a bigger business than say AWS. Gaming is really big, I believe you have heard about gaming being bigger than movies & music combined.

The Crypto market cap is $1.6T, a lot of these Crypto currencies use graphics cards to mine their currencies. So I don't think $200B is too much. For Nvidia which is well positioned in these industries, i.e., owning 80% of that market.

Humbert H. adds:

One fundamental point about predicting the future of NVIDIA. It's a complete accident (lucky for NVIDIA) that the hardware optimized matrix multiplication used for 3D graphics pipelines was also useful for AI.

K. K. Law riffs on The Great One:

Confirmation bias. And this is where the AI computation puck is at of course.

Cagdas Tuna realizes:

Now I see why everyone chasing this momentum with FOMO as all assumptions based on Nvidia will get all of the cake in the market!

Feb

5

Kim Zussman offers:

Meet the Investors Trying Quantitative Trading at Home

Pietros Maneos trades stocks like many of Wall Street’s most sophisticated operations: running dozens of computer-driven strategies in parallel to chase market-beating returns. But he isn’t some tech-savvy math type. He is a published poet who doesn’t know how to code. Maneos, 44 years old, uses online-trading platform Composer.trade to build, test and bet on quantitative trading algorithms that buy and sell stocks and exchange-traded funds out of his home office in Boca Raton, Fla. One algorithm, for example, holds a triple-leveraged exchange-traded fund tracking the Nasdaq-100 index if the S&P 500 index has recently trended higher—and Treasury bills otherwise. He is currently running 72 such schemes he constructed with the application’s graphical interface, but can also type requests in plain English that Composer’s AI will translate into code. “It’s like having my own personal black box,” he said. “You could argue that I’m a hedge fund with 72 strategies.”

Big Al is puzzled by this bit from the above:

Many users praise its simplicity. But several warned about the tax implications of wash sales and the absence of some common Wall Street risk-management tools, such as one that would automatically exit a strategy when a specified loss is reached.

Huh?

Zubin Al Genubi wonders about market microstructure:

On CME is not clear. Is there somewhere how price changes is explained? Seems the asks should go to 0 before price clicks up but they don't. There is a lot of juggling in the queue as well, spoofing, stuffing. I'm reading Flash Crash, by Liam Vaughan.

Jeff Watson responds:

Here is an excellent perspective on spoofing.

Big Al adds:

This book gets recommended a lot but I haven't read it. Pubbed in 2002.

Trading and Exchanges: Market Microstructure for Practitioners, by Larry Harris.

Asindu Drileba recommends:

I am currently enjoying this biography of Jessie Livermore by Patrick Boyle. It's so well narrated, I hope some of you enjoy it.

Henry Gifford observes:

Patrick Boyle says he used to work for Vic.

Jan

28

This is my favourite channel an YouTube. And I liked this particular episode so much it may be my favourite so far:

What The Prisoner's Dilemma Reveals About Life, The Universe, and Everything

The prisoners dilemma is a choice participants need to make that are as follows:

1. If both participants cooperate, they both get $10 each.

2. If only one of the participants cooperate, the defector gets $1, and the one trying to cooperate (be honest) gets $0.

3. If both participants defect (both are dishonest to each other), they both get $1, which is way less than the $10 they would each get by both cooperating.

These are the only four possible states or outcomes of the game. The objective is simple, if the game is repeated for several rounds, under different environments (varying ratio of cooperators & defectors). What strategy should one choose to make the most money? Several agents choose independent strategies and play against each other with whatever strategy they have chosen. All with the aim of making the most money. It turns out that the best strategy for this game amongst different agents is one they call "Tit for Tat". It can be summarised as, "Be Nice, Try to forgive, But don't be a doormat/push over."

Stefan Jovanovich writes:

Pinched from a Stanford course catalog from 1998/9: Axelrod's Tournament:

In 1980, Robert Axelrod, professor of political science at the University of Michigan, held a tournament of various strategies for the prisoner's dilemma. He invited a number of well-known game theorists to submit strategies to be run by computers. In the tournament, programs played games against each other and themselves repeatedly. Each strategy specified whether to cooperate or defect based on the previous moves of both the strategy and its opponent.

Big Al adds:

The Evolution of Cooperation, by Robert Axelrod

We assume that, in a world ruled by natural selection, selfishness pays. So why cooperate? In The Evolution of Cooperation, political scientist Robert Axelrod seeks to answer this question. In 1980, he organized the famed Computer Prisoners Dilemma Tournament, which sought to find the optimal strategy for survival in a particular game. Over and over, the simplest strategy, a cooperative program called Tit for Tat, shut out the competition. In other words, cooperation, not unfettered competition, turns out to be our best chance for survival.

Kim Zussman gets biological:

Cooperation and Darwin:

Cumulative exposure to paternal seminal fluid prior to conception and subsequent risk of preeclampsia

Humbert H. comments:

The original prisoner’s dilemma was about literal prisoners who didn’t get to play even twice with the same “partners”. There are a lot of situations in the real world that map to the prisoner’s dilemma, but a lot fewer that map to playing the same game with the same partners who are rational and capable of learning.

Big Al appends:

Yale Game Theory Course (24 videos), with Dr. Benjamin Polak.

Peter Grieve goes deep:

I am convinced that the principal functions of a healthy society are (1) to get to the good payoff of the Prisoner's Dilemma, and (2) to find an acceptable solution for the Trolley Problem.

Jan

27

Asindu Drileba writes:

A lot of Bitcoiners are expecting a crazy bull run incoming. Their conjecture is that after the halvening, a shock of supply in BTC will cause the price to sky rocket. Previous bull runs have followed this halvening event. It is very refreshing to see a completely different original opinion.

Sam Johnson asks:

You certainly don't need to reveal the source or methodology of the red line data from your timely bitcoin forecast if you don't wish. But when choosing cycles to forecast markets, is there consistency in the order in which you approach finding good cyclical indicators? Do you begin by "chart matching" or finding a leading indicator that visually/numerically correlates well and front-runs certain markets, or do you start with a hypothesis, testing, and then using or discarding such forecasting cycles?

Larry answers:

The forecast here is really simple: it’s just the longer-term cycle forecast for GBTC. I arrive at it by doing a complete cycle search the meld together the 3 with the highest fit.

Andy Aiken asks:

How do you account for the fact that GBTC was a closed-end fund trading at a discount for the past several years, but the discount closed prior to it recharacterizing as an ETF on Jan. 11? This is a one-time event specific to GBTC, not subject to a cycle. What is the significance for bitcoin?

Larry answers (again):

I just use the back-adjusted data as provided.

Andy Aiken adds:

Speaking of mining rewards, the next halving (in which future mining rewards are cut in half, resulting in less reward from mining as well as less inventory to be sold by miners interested only in cash flow), is in about 100 days. This has been historically a (bullish) tailwind.

But with GBTC being converted to a spot ETF, several bankrupt entities are selling their inventory. FTX is now finished selling about $1B in GBTC since Jan. 11, but 3AC has yet to start selling, and that firm had more on its books than FTX. While I am more bullish than your projection, it's interestingly contrarian and would screw with traders' heads as markets like to do.

Jan

26

Daily sd's 1 (1,1,1,1,1,0,0) mean variation .71 PL 2
Daily sd's 2 (0,0,0,0,0,0,5) mean variation .71 PL -18
Correct forecast, but went bust anyway, due to lumping of volatility.

Asindu Drileba asks:

What would be the best strategy to capture the return of this distribution? How would the position size be computed? Say you have $10.

Zubin Al Genubi replies:

OTM option? Don't know which direction so maybe a strangle? Its an example of a fat tail event surprising someone expecting a certain variance. Like the LTCM guys. $.20? 2%? As a hedge. Depends if its hedge or a trade.

William Huggins comments:

what you're picking up on is that variance alone doesn't describe non-normal distributions very well - you need additional tools like skewness (possibly kurtosis) to pick up on those differences. despite having a better description though, there is the presumption that the data generating process is stable across the sample period, and going forward. I've generally found (despite my poor timing record) that money is to be made when the distribution is changing, not stable (the computers rule those waves imo) so detecting breaks may be more valuable than fixed descriptions.

Peter Ringel writes:

I can confirm this from the math-undereducated trading side. Stability is boring, and boredom can lead to undisciplined trades. Shocks and short-term exaggerations are great.

Art Cooper points out:

Stability is boring, and boredom can lead to undisciplined trades. It's Minsky's Theory when this becomes widespread.

Zubin Al Genubi responds:

Thank you Dr Huggins. That is indeed the point that variance, regression, sd, means, should be used with power law distributions with extreme caution or not at all.

Hernan Avella questions:

Why is all that mumbo necessary when all you need is good entries and good stops? The house never closes and there are so many opportunities ahead. f you need that big of a stop, or it gets triggered so frequent that ruins the profits, your system sucks! It’s not a stop-loss problem.

H. Humbert comments:

I think he is saying the system did suck because it relied on improper statistical analysis, using gaussian distributions for prediction when it should have used a more sophisticated statistical analysis that doesn't make such assumption. If you know of good entries reliably without using statistics, more power to you! And maybe he needs volatility swaps in addition to variance swaps and then his system will be A-OK because that could be a simple way to hedge the fat tails. Since I don't trade, I'm just trying to interpret what's flying by.

Humbert H. writes:

Var swap vs. vol swap would be the purest expression. You could also buy a call on realized variance, by buying an uncapped variance swap and selling a capped variance swap (for historical reasons, the cap is struck at 2.5x the variance swap strike, the cap level acting as your effective call strike).

For 100k vega notional and uncapped strike at 22, and capped strike at 20, and realized vol over the period of 80:

100,000/(2*strike) = var notional = 2,272.72 var units uncapped, 2500 var units capped
Pnl uncapped 13.4mm
Pnl capped -4.1mm
Net 9.3mm for ~0.2m cost, not bad (approx (22-20) * vega not).

Some payouts were on the order of 2000:1 during March 2020. Pre 2020 you had some active sellers:

‘Amateurish’ Trades Blew Up AIMCo’s Volatility Program, Experts Say

H. Humbert responds:

Interesting. And an interesting article. You'd think that after LTCM people would realize that 100 year floods are just named that for convenience. That's why I never buy stocks in insurance companies. He whose name shouldn't be mentioned (not the fractalist but the Middle Eastern guy) always advocated buying black swan options, but I think the Chair didn't think he made money on this.

Kim Zussman links:

The hedge fund titan who’s been watching for ‘black swans’ for decades says the ‘greatest credit bubble in human history’ is set to pop—but he’s not worried

Jan

25

People have said that the reason fundamental physics has slowed down is that we have picked all the lower-hanging fruit, but that's not true. There is more lower-hanging fruit than ever before, it's just that picking it is stigmatized.

- David Deutsch

The full podcast is here.

This reminds me of what Brian Arthur insinuated in his book, The Nature of Technology. Brian Arthur describes technology as a combination of other technologies. An example is smart phone being a combination of battery technology, wireless communication technology, a microprocessor technology etc. A common statement I hear often is that we will not see much more technological progress because all the lower hanging fruit (or important things to be invented) are gone. Brian Arthur in his book asserts that if technology is a combination of other technologies, then the invention of new technology should increase the possible space of new technologies that can be invented. For example an AI breakthrough (the invention of the Transformers Model that underlies ChatGPT) will make it easier to invent new products, discover new phenomena which will also make it easier to produce even newer technology. Could this insight be a a good conjecture for always being long technology companies, since we expected technology to grow almost boundlessly if this is true?

Peter Saint-Andre comments:

Although it's seemingly true that technology always grows, that doesn't necessarily mean that technology companies are always a good investment. Various technology industries (crypto, Internet, semiconductors, chemicals, automobiles, radio, railroads, etc.) have experienced cycles of over-investment and hype. I worked in Internet tech companies from 1996 through 2022, and plenty of the companies I worked at either went bust (returning nothing to the investors or employee stockholders) or never approached their former highs (can you say Cisco?). It's not clear to me that, on balance, technology companies provide above average returns. But my perspective is qualitative, not quantitative.

Zubin Al Genubi responds:

That is the Lucretius Fallacy. Thinking the prior highest or best is the top. There will always be something new, bigger, better. That is why NQ is good over time. The old fades out and the new rises ever higher.

Asindu Drileba replies:

It is true that most tech companies actually fail without ever yielding a profit. How ever if your are diversified i.e have a very broad portfolio of investments. You don't have to be successful very many times. You can do very well with a 90% failure rate. Fred Wilson (of Union Square Ventures) claims that half of all VCs beat "The Stock Market" (I am assuming he means the S&P 500).

Big Al writes:

Important, too, to notice the improvements in ordinary things we might otherwise take for granted. A lot of this progress happens in basic materials. A quick search produces:

9 Material Discoveries that Could Transform Manufacturing

During Covid, our dishwasher broke. It was at least 35 years old and possibly older (amazing the use we got from it!). Because seemingly everybody was remodeling while they were stuck at home, it took us 3 months to get a new Bosch (during which time I washed a *lot* of dishes). But I was amazed at what an improvement the new Bosch machine was: it's so much more efficient, with energy and water, and effective, as well as quiet and very smart. That experience woke me up a bit to how much things get improved, and without any central planning authority being responsible for it.

Hernan Avella warns:

Yet, the new Bosch won't last 1/2 of the old one.

Jan

22

This is the best explanation I have seen so far concerning the Poisson Process & Poisson Distribution. It has clearly defined math variables (something explanations involving maths seldom do) & very clear practical examples. I wish more people describing math concepts wrote like this.

A Poisson process is a model for a series of discrete events where the average time between events is known, but the exact timing of events is random. The arrival of an event is independent of the event before (waiting time between events is memoryless).

Zubin Al Genubi comments:

Seems useful to study occurrences of crash or bear market.

Big Al offers:

3Blue1Brown does some great math videos, eg:

Binomial distributions | Probabilities of probabilities, part 1

H. Humbert is skeptical:

It's hard to know without a lot of study whether this is useful for any real-world applications. This distribution has been used in network traffic modeling since the advent of networks because networks have packets and packets have rates that COULD be pretty stable over the period of interest. It worked pretty well for legacy telephone networks, but not so much as computer networks become more and more complex. People still like it because it's a relatively simple formula where if you know the lambda you know everything, and it has no memory of the past so you don't need to store the past, but it doesn't really work well. It doesn't even work that well for predicting meteor showers because the rate itself is subject to change, so can it really work well as a predictive tool for the markets?

Andrew Moe writes:

Poisson has shown to be useful in predicting soccer and hockey scores. In the markets, one test might be to model uncorrelated markets against each other in a double Poisson, like the soccer quants do. Offense and defense, up markets and down.

Jan

17

Statistical Consequences of Fat Tails

Taleb discusses how fat tails can affect probabilities. Is a 10 sigma event an outlier or is it part of a different power law distribution. How slowly does the Central limit theorem conform say Student T distribution to normal (need n>120) for proper confidence levels. Learned about Pareto and other power law distributions. Book suffered from poor editing, missing color references, and Taleb's abrasive pedantics. Recommended nonetheless.

Asindu Drileba writes:

I have learned a lot from both Vic & Taleb. Vic introduced me to obscure trading psychology & insights. Via his books, interviews and books recommended (Horse Trading, Secrets of Professional Turf Betting etc). At first these recommendations seemed strange. But after watching this video by D. E Shaw, it finally made sense because Shaw hints that successful models are built often via thinking in terms of analogies. So, reading Vic's own books, book reccomdations and thinking in terms of analogies can allow you to develop new insights into the market.

Taleb introduced me to the complexity theorists (Didier Sornette, Ole Peters, Mandelbrot etc). He also actually introduced me to Vic's work. Education of a Speculator is praised & highly recommend in Taleb's Fooled by Randomness. I also like Taleb, because he simplifies his concepts in to plain English. So a lay man like me can easily understand what he is trying to say. For instance majority of a statistical consequences of fat tails is summarized in Extreme events and how to live with them- The Darwin College Lecture. In plain English.

Zubin Al Genubi reponds:

The gist of the papers is that use of Gaussian underestimates tail events. Its a good point. Since so many do, it opens good trading opportunity. I've found several which is left as an exercise for the reader and explained in the references here.

Jan

14

I have an interest in prediction markets (also known as information markets or idea futures), such as election betting odds, that allows people to place bets on who they think will be the next president. I wrote an article on my blog some time back (2020) describing the phenomena referred to as the "wisdom of the crowds" that makes these prediction markets possible:

For years now I have been fascinated by prediction markets. The source of excitement is the idea is that you can use financial markets to do inference — just like machine learning. A famous example of such prediction markets are the orange futures. The orange futures market is one that allows entities to buy oranges in advance. How it works, is that one can pay $1,000 to receive 1,000 oranges that will be delivered next year. An interesting side effect of this orange futures market is how it accurately predicts temperatures in certain locations more specifically, the temperature of the locations where the oranges are from.

Peter Ringel writes:

this is a clever thought, and also a terrible situation. I too noticed that it seems - in places - to be easier to predict pockets of the real economy with the financial markets. Of course, traders like it the other way around. Mkts got so efficient. The outside world has way more inefficiency left. (Also enjoyed your mention of "J" language - never heard about it before.) the source of excitement is the idea is that you can use financial markets to do inference.

Zubin Al Genubi comments:

The difference between prediction markets and financial markets is that prediction markets are binary outcomes and markets have non binary outcomes. The distributions are different.

Larry Williams responds:

What a great point. That’s a massive difference….then add in position size.

H. Humbert writes:

An option price seems awfully similar to a prediction market price: both deal with a discrete event at a particular time in the future (or at least close enough for most prediction markets), and right before expiration both, in a way, create a binary choice. I don't trade options, but that's what it appears like.

Zubin Al Genubi replies:

One big difference is options are subject to arbitrage. The prediction markets are not and get wildly inaccurate swings.

Big Al offers:

Binary Option
Superforecasting: The Art and Science of Prediction
Brier scores

From an interview with Michael Mauboussin:

…when you have an investment thesis to buy or sell something, that means you believe you're going to generate an excess return, or there's a mispricing in the market. And…that thesis should have sub-components to it that will allow us to create a scoring system. The most common of these or known of these is called a Brier Score….To have a Brier score you only need three things. You need an outcome that we can agree upon, within a time period that we are finite, with some probability….And so my argument is break down your thesis and put it into some Brier score ready predictions…what I find is the very discipline of writing those things down will force you or compel you to think more…deeply about them. For example, if you're assigning probabilities, you're going to immediately start searching for base rates.

« go back —

Archives

Resources & Links

Search