2
$\begingroup$

Possible Duplicate:
How much data is needed to validate a short-horizon trading strategy?

Suppose I have daily returns for a trading strategy against one month of data. Before starting trading on small quantities or buying further data, I would like to understand the worth of any such action. I am basically looking at the returns per day and the standard deviation between these returns.

  • Is there a standard theory that tells me about the minimum number of data points based on these factors i.e. daily returns, mean return and standard deviation? A minimum number of data points that gives me some sort of a confidence range to invest further for data.

  • Is there a standard theory that lets me infer some results into the future based on a standard approach?

$\endgroup$

marked as duplicate by Tal Fishman Jul 26 '12 at 15:14

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • $\begingroup$ You can't infer the future based on past results; that's just one of the harsh realities of backtesting. As for how much data is needed for a reliable backtest, see How much data is needed to validate a short-horizon trading strategy? $\endgroup$ – chrisaycock Jul 26 '12 at 13:33
  • $\begingroup$ @chrisaycock I agree. However, I guess I did not put my question clearly. I basically wanted to estimate the likelihood of extreme returns because of my sample data being limited. I found the following discussion on the topic: quant.stackexchange.com/questions/3839/… And thanks for the link, it answers my question. I guess that renders this post as "redundant" :) $\endgroup$ – Aziz Jul 26 '12 at 14:47
  • $\begingroup$ @Aziz No problem, glad you found your answer. Your post has been closed as a duplicate, but feel free to post new questions after you've looked around to see what we already have here. $\endgroup$ – Tal Fishman Jul 26 '12 at 15:15