Suppose I had a statistic for which there was no published critical values, nor distribution. How are distributions for such statistics typically calculated?
I have been using the Spearman-Conley statistic (used in hydrology, similar to Spearman statistic) and there are critical values published for sample sizes up to n = 30. I am interested in developing critical values for larger sample sizes.
My intended approach is to take many, many random samples and calculate the statistic for each of the random samples and thus develop a frequency distribution for the statistic.
I am curious how it is usually done.
Advertise on Talk Stats