Statistical Significance & Results

[runningman]

If I know my average win and standard deviation of returns, how can I find if my results are statistically significant or just the result of chance?

 

[dtrader98]

Quote from runningman:

If I know my average win and standard deviation of returns, how can I find if my results are statistically significant or just the result of chance?

If you have a lot of observations and a fairly normal distribution, one method is to try generating a random run of returns with avg win of zero and equal std dev, then run a t-test (freely available in excel).

Good trading type book on the subject is Aroonson, Evidence Based TA.

 

[puretruth]

Quote from dtrader98:

If you have a lot of observations and a fairly normal distribution, one method is to try generating a random run of returns with avg win of zero and equal std dev, then run a paired t-test (freely available in excel).

Good trading type book on the subject is Aroonson, Evidence Based TA.

Is that person a trader himself ?

 

[NanoTick]

Quote from runningman:

If I know my average win and standard deviation of returns, how can I find if my results are statistically significant or just the result of chance?

google "confidence interval"

 

[dtrader98]

Quote from puretruth:

Is that person a trader himself ?

A little research will reveal his background for you.
http://www.evidencebasedta.com/aboutAronson.html

 

[ivanbaj]

This could help you:

http://www.elitetrader.com/vb/showthread.php?s=&threadid=33654&perpage=6&pagenumber=1

Make sure you read page 5 and then decide.

 

[Craig66]

This may also help:

http://www.elitetrader.com/vb/showthread.php?s=&threadid=36083

 

[runningman]

Quote from Craig66:

This may also help:

http://www.elitetrader.com/vb/showthread.php?s=&threadid=36083

perfect- that's exactly what I was looking for. Looks like this is the formula I wanted:
x = abs(Avg-Profit) * (Number of Trades)^0.5 / (Std Dev of Profits)