Z-Ratio


Z-Ratio

The Z-Ratio is the result of a Z Test of Statistical Significance.  While the number does not identify how much better than random the trades produced from the strategy are (i.e. importance), the Z-Ratio is helpful in identifying whether the trades are significantly different (i.e. better) than trades produced randomly for the same period.  The higher the Z-Ratio, the lower the probability that the trades were produced by random chance.  Statisticians often use 3 critical ratios to determine statistical significance.  These are:


Alpha Level

Critical Ratio

.05

1.64

.01

2.33

.001

3.09


If the Z-Ratio is greater than 2.33, then the trades can be said to be statistically significant at the alpha level of .01.  This means that there is only a 1% probability that the trades were produced by random chance.  If the Z-Ratio exceeds 3.09, there is only a 0.1% chance that the trades were produced by random chance.  On the other hand, if the Z-Ratio were less than 1.64, most statisticians would assume the probability is too high that the trades were produced randomly, and would therefore consider the results statistically insignificant.


Statistical significance is an important factor in determining the quality of trading systems. StrataSearch uses the Z-Ratio, along with many other performance metrics, to identify profitable trading systems. For more information on StrataSearch, click here.


Note 1: A negative Z-Ratio indicates the average trade selected randomly had a higher return than the average trade produced from the strategy.


Note 2: Since a different sampling of random trades is produced for comparison each time the strategy is run, the Z-Ratio may differ slightly between runs.