|Posted on January 6, 2011 at 1:00 AM|
Over the 10 years of data on which the system was optimized, there were 45 trades (n= 45). The mean or average trade yielded about $199.4936, and the trades were normally variable, with a sample standard deviation of around ±$150.7709. The expected standard deviation of the mean suggests that if samples of this kind were repeatedly taken, the mean would vary only about one-tenth as much as the individual trades, and that many of the samples would have mean profit abilities in the range of $199.4936 ± $22.47549.
The t-statistic for the best-performing system from the set of optimization runs was 8.876, which has a statistical significance of 0.000. This was a fairly strong result.
The serial correlation (Durbin-Watson) was 1.552, while the dL value is 1.48 and the dU 1.57. These results strongly suggest that there was no meaningful serial correlation between trades and that the statistical analyses discussed above are likely to be correct.
Sometimes we need to make sure that the system is really makes profit. To realize it we used the 25% percentiles as test value.
THE ENCYCLOPEDIA OF TRADING STRATEGIES
JEFFREY OWEN KATZ, Ph.D.
DONNA L. MCCORMICK.