Stata 15 help for postest
[R] postest -- Postestimation Selector
Syntax
postest
Menu
Statistics > Postestimation
Description
Launch the Postestimation Selector window. The window contains a list of
all postestimation features that are available for the currently active
estimation results. To launch the dialog box for an item in the list,
select an item and click on Launch. The list is automatically updated
when estimation commands are run or estimates are restored from memory or
disk.
Remarks
Stata uses an estimation followed by postestimation analysis paradigm.
You type regress ... to fit a regression model, then you type test ... to
test linear relationships among the estimated parameters, or you type
contrast ... to compare marginal means, or you type rvfplot to see a
residual-versus-fitted plot, or you type one of a myriad of other
postestimation commands. This is an extension of Stata's "type a little,
get a little" concept. The Postestimation Selector exposes this type of
postestimation analysis to those who prefer to use the dialog boxes to
fit and analyze models, or at least they sometimes prefer the dialogs
when exploring their data. We might call this "click a little, get a
little".
The Postestimation Selector knows what is available after any estimation
command. It shows the full list of postestimation features that are
available after any estimation, and it shows only those that are
available. For example, if you fit a linear regression, you can choose
from 59 postestimation analyses, including "Likelihood-ratio test
comparing models". If, however, you fit that linear regression using
survey estimation, the likelihood-ratio test is not available because
that test has no meaning for survey estimation. You will, however, see 8
new postestimation features for survey-data analysis, including "Design
and misspecification effects".
If you are using the menus and dialogs, we recommend you launch the
Postestimation Selector and just leave it up -- select Statistics >
Postestimation, or type postest at the command line. All available
postestimation features for whatever model you are analyzing will then be
just a click away.
Video example
Postestimation Selector
|