Dear STATA Users,
I have a question about the nested logit model.
I will first explain my situation, and second address my question.
I will run the nested logit model.
My dependent variable (DV) has four sub-categories which are nested in two
big categories:
Category1:
sub-category11
sub-category12
Catergory2:
sub-category21
sub-category22
I have several independent variables (IV): IV1, IV2, IV3, IV4, IV5, IV6,
IV7, ...
I assume that IV1, IV2, IV3 will not only influence category1 vs. category2,
but also influence the type of subcategory.
Question: Could I repeat IV1, IV2, IV3 in the nlogit model, as follows?
nlogit chosen ||category_1_2: IV1 IV2 IV3, base(1) || type: IV1 IV2 IV3 IV4
IV5 IV6 IV7, case(id)
Is it correct to repeat several independent variables in predicting both the
large categories and the sub-categories?
Thank you very much.
Sincerely,
Patricia
*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
PATRICIA YU
Ph.D. Student & Project Assistant
Dept. of Educational Leadership & Policy Analysis
University of Wisconsin-Madison
Cell: 608/698-8108
Home: 608/231-6855
E-mail: [email protected]
*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
