I would like to test the significance of difference between the coefficients in 2 probit models. I follow the code in Stata Manual:
. logit y1 x1
. estimates store M1
. logit y2 x2
. estimates store M2
. suest M1 M2
. test [M1]x1=[M2]x2
In my case, I re-phrased the above code:
. probit y1 x1 if group==0
. estimates store M1
. probit y1 x1 if group==1
. estimates store M2
. suest M1 M2
. test [M1]x1=[M2]x1
But I was only told by Stata that "equation M1 not found" so that I have to use "reg" model instead of "probit" to test the coefficient difference.
I am really thankful if anyone could kindly provide suggestions on the proper stata code. Thank you very much!
Related Posts with Testing coefficient difference in 2 probit models
Fitting Poisson DistributionDear all, I have a variable - number of people who are malnourished in a PSU. I have approximately 2…
The matter of separating probit and truncated regression in churdle commandHi, I'm new here. I'm currently using churdle command to implement double hurdle model, and I found…
Palma ratioHello I have the following information How can I get the Palma ratio for each province in Iran? (Eac…
Robust option and cluster option used together in regressionIs it appropriate to use the robust option (Huber-White estimator) and cluster option together? For…
How to average duplicate variables so that I can reshape dataHey guys, This is the data I have - Code: * Example generated by -dataex-. To install: ssc insta…
Subscribe to:
Post Comments (Atom)
0 Response to Testing coefficient difference in 2 probit models
Post a Comment