Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Oredered probit vs. Mixed model

From   "ALICE DOBSON" <[email protected]>
To   [email protected]
Subject   st: Oredered probit vs. Mixed model
Date   Sun, 24 Jul 2005 08:05:45 -0400

I use Stata 9 and have a query on Mixed model.

I have an unbalanced panel of entry by firms into various towns (M=1123) in a particular industry in the US during the period 1969-1976. The number of firms that enter each town during each year varies from 0 to 39. Two different types of firms enter: (a) those who sell technologically sophisticated products, and (b) those who sell technologically inferior products. The variable SUPTECH denotes the number of firms which offer technologically superior products while INFTECH DENOTES those supplying technologically inferior products. The data structure looks like this:

M1 1969 1 0 0
M1 1970 1 1 0
M1 1971 2 0 0
M1 1972 2 4 0
M1 1973 6 5 3
M1 1974 11 3 2
M1 1975 14 1
M2 1969 3 0
M2 1970 3 1
M2 1971 4 0
M2 1972 4 4

In addition there are a number of other independent variables such volume of sales in each town (market size), demographics, and so on. I am testing several hypotheses concerning the nature of entry. However, my query to statalisters is a simple one.

I am testing whether the probability of entry of firms offering inferior technology products increases with the number of firms. I use an ordered probit model for this where the dependent variable is INFTECH (0 to 14). However, since I have a panel data could I also use a mixed model?


Express yourself instantly with MSN Messenger! Download today - it's FREE!

* For searches and help try:

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