Feb

27

Wouldn't the adjusting up of the prior contract data to the current destroy information about the beneficial effects of inflation on stocks and owning assets?

Leo Jia responds:

I had considered about this and believe the adjusting does destroy some information, but one can go around the problem.

Generally I use one of two schemes to adjust: subtraction, or division, each destroying the info in a different way. Which one to use depends on one's analytical formulas to be used. For instance, if one is concerned about absolute price differences, like close of today minus close of 2 days ago, then one needs to use the subtraction scheme; one the other hand, if relative difference is of concern, like close of today divided by close of 2 days ago, the division scheme should be used. Using it the right ways nullifies the information destruction.

The subtraction scheme can produce an artifact of prices becoming negative, so mostly I concentrate on the division scheme.

Btw, I open-sourced the adjusting routine called Stitcher (in Julia) on GitHub.

Steve Ellison adds:

Much depends on what you are trying to achieve by using adjusted prices. I use them to make sure my calculations of net price changes and n-day highs and lows are accurate in the event such calculations cross a contract roll. When back-testing, I typically do selection using the adjusted prices and then translate the specific occurrences back into the contract that would have been used at the time of entry– then I can compare the net change to the original price, resulting in a more meaningful percentage change.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search