estat
The estat
command calculated scalar and matrix based statistics. The precise capabilities of
estat
depend on what command you run before it. There are three types that are available after any estimation command. For specific forms of estat
available after a command, try typing help [command] postestimation
For example. help regress postestimation
or see the corresponding entry in [R]. The three postestimations performed by estat
following any estimation are estat ic
estat summarize
and estat vce
are are the focus of the below.
The estat ic
command calculates two information criteria (ic) which can be used to compare the fit of different models. It can be used after any command which includes a report of log likelihood. The first criterion computed is the AIC short for Akaike Information Criterion. A basic introduction and links to useful resources can be found on Wikipedia. The second information criterion computed is the BIC or Bayesian Information Criterion. A basic introduction and links to useful resources can be found on Wikipedia. These two criteria are not always in agreement about which model is a better fit.
The estat summarize
command shows summary staistics (mean, std. deviation, min, and max) for all variables involved in the model. Additionally, you can specify ,labels
to display variable labels
noheader
to suppress the table header noweights
to ignore weights from a previous estimation command and equation
which displays repeat summary statistics multiple times for variables involved in multiple rquations.
The estat vce
command calculates the variance-covariance matrix of the parameters of the model used in the estimation command. A fairly basic intro to covariance matrices can be found here. Another basic introduction can be found at the government's NIST Statistics for Engineering page. A good explanation/example of variance and covariance can also be found from the Visual Statistics website.