ssm
by Alfonso Miranda and Sophia RabeHesketh
Models
ssm is a wrapper for gllamm to estimate endogenous
switching and sample selection models for
binary, count, and ordinal variables by maximum likelihood using
adaptive quadrature.
The model consists of two submodels:
the outcome model and the selection or switching model.
For binary or ordinal outcomes, the outcome model
is a logit or probit model; for counts the outcome model
model is a Poisson model. If the problem is an endogenous switching
problem, the switching variable appears as a covariate in the outcome model.
The switching or selection model is a probit model.
The commands option causes ssm
to print out all data manipulation commands and the
gllamm command for estimating the model. gllamm itself can then
be used to extend
the model or to make predictions or simule data gllapred
or gllasim .
Reference: Miranda and RabeHesketh (2006).
Maximum likelihood estimation of endogenous
switching and sample selection models for
binary, count, and ordinal variables. The Stata Journal 6, 285308.
Installation
The command requires Stata 9 or later (available from Stata Corporation)
and the latest version of gllamm (see here for
installation instructions).
ssm can be installed from Statistical Software Components (SSC):
. ssc describe ssm
. ssc install ssm
. ssc install ssm, replace /* to replace previous version */
or downloaded directly from here: ssm.ado, ssm.hlp
