Message:

<p>[QUOTE=&quot;medoraman, post: 1729618, member: 26302&quot;]Its called backfiting a model in Finance. Its used all of the time to make &quot;market predictions&quot; to suckers. Its also the bane of a Finance professor&#039;s existence. I would markedly warn my students if I judged their models to be backfitted they would fail. </p><p><br /></p><p>The problem with backfitting, (basically taking number history with no other cause inputs), to predict the future is the numbers are the RESULTS of a million inputs, not the CAUSES. This is my problem with most technical trading, they act as if the numbers are creating the other numbers. Maybe they do short term, (myriad of reasons), but long term analysis has to be performed on the real CAUSES of market changes, not the scorecard of the results. </p><p><br /></p><p>If the Yankees and Mets had a three game series, do you think the scores of the first two games will predict exactly what the score of the third game will be? Don&#039;t you think you want to know who&#039;s pitching, if there are changes to the lineup, etc? How about the history of Yankees and Mets from the last three years? How much predictive value do those scores have to a game played tomorrow?</p><p><br /></p><p>I can make a model, (I actually did it), incorporating precipitation in Iowa to help &quot;predict&quot; the average wholesale price of green tea in Beijing. It was a tongue in cheek, literal &quot;What does that have to do with the price of tea in China&quot; demonstration. Point is, ANY PRICE can be &quot;correlated&quot; with something else, or itself, to &quot;predict&quot; something. Is it of any use? Not really, but you will see tons of &quot;leading industry experts&quot; flouting their &quot;proprietary models&quot;. </p><p><br /></p><p>Just a general warning. Not trying to pick on anyone in particular.[/QUOTE]</p><p><br /></p>

Your name or email address:

Do you already have an account?

• No, create an account now.
• Yes, my password is:
• Forgot your password?

Stay logged in