Clarification of Question by
gohou-ga
on
22 Nov 2004 07:56 PST
Dear ,
I would like to run a multinomial Probit and Nested logit with SAS.
The data are such as for each observation, I have one possible choice
(the number of alternative for the choice is 4 for level 1 and 2 for
level 2) and the dependant variables are related to the
observations, not the alternatives. Here is an example of the data
id Choice _top choice_bottom age education ...
1 1 1 45 3
2 1 2 56 6
3 2 1 39 4
The top choice has 2 alternatives and each alternative of the top
choice has 2 alternatives. At the end, I have 4 alternatives.
The dependant variables are link to the observations, not the
alternative.
I transform the data before the procedure MDC:
id Choice _top chosen age education choice ...
1 1 1 45 3 1
1 1 0 45 3 2
1 0 0 45 3 3
1 0 0 45 3 4
2 1 0 56 6 1
2 1 1 56 6 2
2 0 0 56 6 3
2 0 0 56 6 3
.....
Below is the code I used. Unfortunately, none of my variables is
significative (when the model converge, which is obvious)
proc mdc data=GOHOU.V2 type=nlogit ;
model chosen = age education
/ choice=(t 1 2 3 4) maxiter=3000 covest=op;
id hid93;
utility u(1, 1 2) = age education,
u(1, 3 4) = age education,
u(2, 1 2 ) = age education ;
nest level(1) = (1 2 @ 1, 3 4 @ 2),
level(2) = (1 2 @ 1);
output out=gohou.res10 p=proba xbeta=beta;
run;
The first setting I have done with SAS and stata doesn't converge and I
don't know what may be the problem.
I have 2 questions:
1. How to set up the data in my case to run this model?
2. How to interpret the result I will get?
Thanks in advance.
