Jan

25

Vic gives out a quest (a research quest).

I swear, I did see Vic's X post only afterwards. I too was thinking about the levels / rounds and how to test them today during my drive. Probably everybody is tired of starring at the same levels and so the mind wonders and wanders.

I am thinking of this approach:

utilize volume profile / volume per price level
(here a buffer bin can be applied / granularity from tick to x points)
(this will become important)
slice / bin the time series in segments of x
x needs to be defined,
I am thinking simple price ranges right now
normalize the segments
create a summary volume profile of all the segments ( an average or total sum )
plot it and hope something stands out
then a deeper statistical analysis of the volume profile, which is a frequency distribution

another approach could be to look at the "rejection power" of a price level after a crossing / touching event.

the crossing / touching events are often fuzzy in time
maybe remove time here a la range bars
after the event qualify the price range traveled for a fix time interval
so price-no-time vs price-time

Zubin Al Genubi writes:

The sample generally is biased bullish. Maybe take a look at bear regimes to see how hypothesis hold up. Also need to use unadjusted prices to retain rounds.

Peter Ringel responds:

TY. I am also thinking cash and virgin levels (around ATHs). the question of prices vs percentages will come up here too again.

If one sticks to prices and assumes an underlying behavioural cause of rounds, there probably are regimes with more or less black and white borders (fast transition) from 50 with weight to 100 with weight to maybe back to 50 in low volatility environments. Also the question, what difference does it make, if the index is 7000 vs 700.

William Huggins suggests:

it may be worth seeing if there are "liquidity cliffs" at rounds rather than sticky prices (order clustering as opposed to price clustering)

Peter Ringel adds:

The open interest at rounds at the option chain is usually noticeable higher vs all other steps.

Now I am thinking to look at the waning of these. Are they getting eaten up. Would explain the multiple crossings of a round needed till we get new ATHs.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search