Closed waynelapierre closed 1 year ago
Hi,
Thanks for your interest in the package. You could try the option ckertype=”epanechnikov” in the call to the bandwidth generation and the call to npsigtest, plus set the global option option(np.tree=TRUE) (see FAQ for a tad more info), and drop the number of bootstrap reps to 99 from 399… see if that helps?
Does dropping the number of bootstrap reps make the result less reliable? Thanks.
Hi, Slight reduction in power, that’s all… size unaffected…
Jeff
Is there any way to make npsigtest faster? I have an R script where more than 95% of the time is spent on multiple npsigtest functions.