Inflation-Adjusted Historical DB?
Author: rmpwealth
Creation Date: 7/5/2011 8:44 AM
profile picture

rmpwealth

#1
Recently I tried to use Wealth-Lab to backtest a strategy my friend read on a financial website. The results were not even close, because the website's trade amounts were based upon "current value", which I interpet to mean that the historical data used was inflation-adjusted.

1. Can someone direct me to a DB that W-L can access that has inflation-adjustd historical price data?

2. Question: Is using inflation-adjusted historical price data considered to be a good idea? Since this is the first time I've encounted the concept, I am wondering why no one else seems to use it (or maybe they do, and I'm just not aware). FWIW, I use Fidelity's DB for most everything, except use Yahoo when I want to use dividen-adjusted prices.

Thanks.
profile picture

Cone

#2
Hey, this sounds interesting. We'd have to make sure the system meets certain criteria, but how would you like to write or co-author a Trading System Lab article for Active Trade Magazine that is based on the system you found on a website?

Re: inflation-adjusted...
For it to be valid, my opinion is that the inflation adjustment would have to come in the form of trade returns over time, so you wouldn't really adjust the historical price data. Sounds like a Performance Visualizer function. Eugene, what do you think?
profile picture

Eugene

#3
Robert,
Help me follow you. Which visualizer? What to adjust? Is inflation a series (then where to take it from) or a fixed number (throughout the simulation)?
profile picture

Cone

#4
Which visualizer? A new one.

Y-Charts has an inflation series. The idea would be to present the trade and other performance returns in inflation-adjusted terms. I haven't thought about it much, but it's a reasonable idea from a wealth-building perspective to know that a 100% gain over a 5-year period that averaged 5% inflation each year isn't even close to a 100% gain in real terms.

rmpwealth can probably help us inflation novices. Would it be enough to adjust Annualized Gain for inflation?
profile picture

rmpwealth

#5
QUOTE:
Would it be enough to adjust Annualized Gain for inflation?


I think that is what the original post's referenced strategy did. According to my friend, the author of that strategy simply took the backtest's percentage return, then applied an inflation factor to it. I think this is what Cone suggested in his revised 07/05/11 post.

After thinking about this subject a little more, it seems to me that inflation would affect all strategies the same. Therefore, while it might be "nice to know" what an inflation-adjusted result might be, I'm not sure that taking inflation into account during a backtest would contribute any additional value when evaluating and comparing different strategies. If you agree, please don't do a lot of work on this on my behalf.

Thanks.



profile picture

Eugene

#6
In the article on Inflation-Adjusted Return, Investopedia suggests a similar approach: "a simple approximation for inflation-adjusted return is given by subtracting the inflation rate from the rate of return."

They also show a more formal approach:

profile picture

streak

#7

And then if your back test is for relevantly long term holds you'd also need to account some difference to the inflation adjusted cost (as opposed to 'return').

Also there might be bonuses distributions to consider? ...and more(?)
profile picture

Cone

#8
QUOTE:
you'd also need to account some difference to the inflation adjusted cost
How so? Change the data in a way that wasn't present in the market conditions based on future inflation?

I think adjusting the returns is right way to go. It gives you the inflation-adjusted value of the returns (today) without changing the historical absolute returns or trades.
profile picture

streak

#9

I made an impromptu comment. Yes not completely thought out or investigated, and take your point.

Was just considering the multitude of variables that could come into play when accounting returns beyond the 'immediate'(syntax?) values. eg things like bonuses/rights taken up (or not), loss or gain from entry cost (accounting inflation on account balances while flat - which could negate an 'inflation adjusted return' could it not?), back testing on split data (often wild ratios here in Aus like 100:1) - and so entry at some mid bar price may have been impossible, and similarly back testing on say eod resolution only that cannot account entry at some price between bar low & high that didn't exist due to an intraday (news) gap, etc.

If one is accounting inflation adjusted returns would it not also be pertinent that many of the other variables be fully accounted?