Dear Statalist,
Upon using weakiv after an ivprobit (first stage is binary and second stage as well), what does it mean when the confidence set is the entire grid ?
Also, I want to make sure that the AR-chi2 is comparable to the F-test ?
Here is what I get as results
------------------------------------
H0: beta[work:married22] = 0
Test Statistic p-value Conf. level Conf. Set
CLR stat(.) = 0.29 0.6924 95% entire grid
K chi2(1) = 0.06 0.8129 95% entire grid
J chi2(1) = 1.89 0.1689 95% entire grid
K-J <n.a.> 0.7303 95% (96%,99%) entire grid
AR chi2(2) = 1.95 0.3774 95% entire grid
Wald chi2(1) = 0.00 0.9814 95% [-4.80667, 4.92228]
Confidence sets estimated for 100 points in [-9.67115, 9.78676].
Appreciate your help.
Irene
Upon using weakiv after an ivprobit (first stage is binary and second stage as well), what does it mean when the confidence set is the entire grid ?
Also, I want to make sure that the AR-chi2 is comparable to the F-test ?
Here is what I get as results
------------------------------------
H0: beta[work:married22] = 0
Test Statistic p-value Conf. level Conf. Set
CLR stat(.) = 0.29 0.6924 95% entire grid
K chi2(1) = 0.06 0.8129 95% entire grid
J chi2(1) = 1.89 0.1689 95% entire grid
K-J <n.a.> 0.7303 95% (96%,99%) entire grid
AR chi2(2) = 1.95 0.3774 95% entire grid
Wald chi2(1) = 0.00 0.9814 95% [-4.80667, 4.92228]
Confidence sets estimated for 100 points in [-9.67115, 9.78676].
Appreciate your help.
Irene
