Hello everyone. Thank you all for taking the time to answer other questions on this forum, it had been very helpful. This is my first time posting here. I would like to ask a question regarding the oparallel and brant test in particular. As you can see, the assumption of parallel lines is violated.
Tests of the parallel regression assumption
| Chi2 df P>Chi2
-----------------+----------------------
Wolfe Gould | 63.74 22 0.000
Brant | 57.02 22 0.000
score | 60.3 22 0.000
likelihood ratio | 63.02 22 0.000
Wald | 58.48 22 0.000
Brant test is also. However, I am confused why doesn't the chi2 table add up to 57.02? It is much less if I add it up on the calculator. I apologize if it is a really stupid question, but I humbly request your help.
Brant test of parallel regression assumption
| chi2 p>chi2 df
-------------+------------------------------
All | 57.02 0.000 22
-------------+------------------------------
region | 0.02 0.881 1
sex | 0.03 0.870 1
age | 0.76 0.384 1
edu | 3.57 0.059 1
sonh | 1.28 0.258 1
dauh | 3.62 0.057 1
2.mainsrc | 0.22 0.641 1
3.mainsrc | 0.12 0.734 1
4.mainsrc | 0.86 0.355 1
savin | 0.33 0.568 1
gold | 1.02 0.314 1
bond | 2.08 0.150 1
house | 0.00 0.952 1
land | 2.11 0.147 1
car | 0.94 0.333 1
2.satisf | 1.43 0.232 1
3.satisf | 0.09 0.760 1
2.health | 0.78 0.377 1
3.health | 3.89 0.049 1
1.templ | 0.32 0.571 1
2.templ | 0.79 0.373 1
3.templ | 0.02 0.892 1
When I run gologit2 there are also no constraints found. Image attached.
Thank you all. I sincerely hope to get some answer.
No comments:
Post a Comment