Postestimation Commands & Regression
After a regression, there is a variety of follow-up work you may want to do. This work is done using posetestimation commands. All a postestimation command is, is a command that can only be run after an estimation command. The regress command is one option among many. However, following regression there are some postestimation commands of special interest. Below is a list of every command which you might want to run following a regression, followed by a brief description of what the command does (taken from [R} postestimation tools for Regression). Click a command to learn more about how to use it and why you might want to.
Special post-regression commands
dfbeta: DFBETA influence statistics.
estat hettest: use this to perform various tests for heteroskedascity (nonequal variances) an assumption of some types of regression.
estat imtest: performs an information matrix test.
estat ovtest: Ramsey regression specification-error test for omitted variables (test for model misspecification).
estat szroeter: Szroeter's test for heteroskedasticity.
estat vif: variance inflated factors for the independent variable.
acprplot: augment component-plus-residual plot.
avplot: added-variable plot
avplots: all added-variable plots in one image
cprplot: component-plus-residual plot
lvr2plots: leverage-versus-squared-residual plot
rvfplot: residual-versus-fitted plot.
rvpplot: residual-versus-predictor plot.
Standard postestimation commands available after regression
adjust: predicted values displayed in tables based on the model
estat: AIC, BIC, VCE, and estimation sample summary.
estat (svy): postestimation statistics for survey data.
estimates: managing estimation results.
hausman: Hausman's specification test.
lincom: point estimates, standard errors, testing, and inference for linear combinations of coefficients.
linktest: link test for errors in model specification.
lrtest: likelihood-ratio test.
mfx: marginal effects or elasticities.
n1com: point estimates, standard errors, testing, and inference for nonlinear combinations of coefficients.
predict: predictions, residuals, influence statistics, and other diagnostic mmeasures.
predictn1: point estimates, standard errors, testing, and inference for generalized predictions.
suest: seemingly unrelated estimation.
test: Wald tests for simple and composite linear hypotheses.
testn1: Wald tests for nonlinear hypotheses.