[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Stata tool for a complex question

From   "Roger H. Coupal" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: Stata tool for a complex question
Date   Tue, 12 Jan 2010 05:46:48 -0700


We are looking for code that estimates a random utility model with three indirect utility equations and 2 latent class variables (LCM) using data from a stated choice experiment. We are using a random utility model for the stated choice (A, B, or neither) and are finding heterogeneity across the survey sample that a LCM may best capture.
Basically the following situation is being modeled:
 - Choice 1, Choice 2, Choice 3 (neither 1 nor 2) which are mutually exclusive choices
 - indirect utility functions are estimated for each choice  of a given choice occasion:
Ch1=f(env attributes, sociodemo attributes)
Ch2=f(env attributes, sociodemo attributes)
Ch3=f(variables, sociodemo attributes differing from Ch1 and 2)
CH 1, 2, and 3 are (0,1) dependent variables and explanatory variables are mix of dummies and actual numbers. 
The likelihood function must take into account the probability of choosing a choice (1,2, 3), conditioned on being within a certain segment delineated by the latent class variable.
 Then a ML function is estimated where the parameters of CH1 and CH2 are the same but the parameters of CH3 are different. Is there a command in Stata that can task this problem? Thanks.

Is there tool in Stata that can accomplish this? thanks

Roger Coupal, Department Head
Agricultural and Applied Economics
University of wyoming
[email protected]

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index