06-04-2012 04:09 AM
In the Durbin-Watson test, the decision rule for a one-sided test is
Reject H0 if d < dL, do not reject H0 if d > dH, dL<d < dH is the inconclusive range.
In the AUTOREG procedure, the DWPROB option produces p-value for the Durbin-Watson test statistic. So what's the decison rule for the p value in a one-sided tes or a two-sided test? What is the inconclusive range for p value?
06-04-2012 08:32 AM
I believe the p values reported address only the rejection part, and do not do anything with the inconclusive range. I hope someone with more familiarity with the many procs that report a DW stat can step in with a more definitive answer.