[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: logit w/interaction of 2 variables
I'm trying to put together a logit model (using survey data) that inlcudes
dichotomous indicator of suicide attempts and an interaction term for
sexuality (0/1)* school environment (continuous) and several control
variables. In particular, I want to be able to test for differences in the
probabillity of suicide attempt at 3 different values for school environment
and to get predicted probabilities.
I'm fine with putting the initial model into Stata form: [note: I'm using
version 8 and this is a simplified model]
xi: svylogit suicideattempt i.sexuality*environment female SES
then, to get odds ratios, i run
then, I want to get a predicted probability for 3 possible values of
"environment", so I generate the following 3 versions of environment
("envlow"=1 standard deviation below the mean for "environment",
"envmean"--at the mean, and "envhigh"--1 standard deviation above the mean)
can anyone tell me how to go once I've generated these variables?
thanks a lot
Mortgage rates as low as 4.625% - Refinance $150,000 loan for $579 a month.
* For searches and help try: