The Motley Fool Discussion Boards

Previous Page

Investing/Strategies / Mechanical Investing


Subject:  Re: maybe done Date:  5/17/2013  5:56 PM
Author:  Radish Number:  243141 of 271050


Hey! Aren't you one of the key architects of the database, backtester, etc??? Say it ain't so!

Well, I wrote a series of Excel macros that simply applying screens to downloaded stock data. The backtesters, and the screen designs themselves, are far more important to MI.

The fundamental problem as I see it is that back testing a screen tells you how that screen worked in a world that no longer exists and will probably never exist again. Unless that screen performed well due to fundamental principals that continue to exist into the future, there's no reason to think that a screen that did well in the past will do well in the future also.

Unfortunately, there's really no way to know whether the principals of a screen's success continue into the future or not. Someone who has an exceptional understanding of everything that affects stock prices will be good at determining which principals matter and which don't... but that "someone" isn't me.

I used to think that you're better off with screens that have worked well in the past than without having any data from the past to guide you at all. These days, I'm not so sure that's true. It limits you to those things that can be backtested.

Copyright 1996-2018 trademark and the "Fool" logo is a trademark of The Motley Fool, Inc. Contact Us