D*********Y 发帖数: 3382 | 1 快要抓狂了。
我现在有三个equations,一个probit,俩tobit,根本没有找到code
发现前人有做过一个OLS,一个TOBIT
或者一个OLS,一个PROBIT。
但是也是没有code可寻。
更不要说像我这种情况了。
想用PROC syslin,全部设OLS,我有fixed effects,该如果code呢?谢谢! |
A*******s 发帖数: 3942 | 2 proc nlmixed seems to be the only way u can do it.
glimmix does not support self defined likelihood function, so u cannot do
tobit.
qlim does not support tri-variate model fitting.
I recall someone posted the code using nlmixed for Heckman model in SAS-L
forum. you may have some ideas after checking it.
【在 D*********Y 的大作中提到】 : 快要抓狂了。 : 我现在有三个equations,一个probit,俩tobit,根本没有找到code : 发现前人有做过一个OLS,一个TOBIT : 或者一个OLS,一个PROBIT。 : 但是也是没有code可寻。 : 更不要说像我这种情况了。 : 想用PROC syslin,全部设OLS,我有fixed effects,该如果code呢?谢谢!
|
D*********Y 发帖数: 3382 | 3 In my equation system, I also have endogenous variables on the right hand
side of the equation. I wonder whether I set up the systems too complicated.
I want to try linear specification now. What is the probelm is I go with
linear specification but not non-linear regressions which is supposed to be
the correct setup.
【在 A*******s 的大作中提到】 : proc nlmixed seems to be the only way u can do it. : glimmix does not support self defined likelihood function, so u cannot do : tobit. : qlim does not support tri-variate model fitting. : I recall someone posted the code using nlmixed for Heckman model in SAS-L : forum. you may have some ideas after checking it.
|
A*******s 发帖数: 3942 | 4 sorry i am not familiar with econometric terms. probably i misunderstood ur
point.
i thought ur previous question is about tri-variate joint modeling...
isn't linear or non-linear another different issue?
complicated.
be
【在 D*********Y 的大作中提到】 : In my equation system, I also have endogenous variables on the right hand : side of the equation. I wonder whether I set up the systems too complicated. : I want to try linear specification now. What is the probelm is I go with : linear specification but not non-linear regressions which is supposed to be : the correct setup.
|
a***d 发帖数: 336 | 5 so you want to find the equilibrium of your equation system?
maybe you can try to estimate coefficients the equations separately as the
first step. Then use other ways (analytical or numerical) to solve the
equation system with estimated coefficients to get the equilibrium.
complicated.
be
【在 D*********Y 的大作中提到】 : In my equation system, I also have endogenous variables on the right hand : side of the equation. I wonder whether I set up the systems too complicated. : I want to try linear specification now. What is the probelm is I go with : linear specification but not non-linear regressions which is supposed to be : the correct setup.
|