help estat dialog: estat
-------------------------------------------------------------------------------
Title
[R] estat -- Postestimation statistics
Syntax
Common subcommands
Obtain information criteria
estat ic [, n(#)]
Summarize estimation sample
estat summarize [eqlist] [, estat_summ_options]
Display covariance matrix estimates
estat vce [, estat_vce_options]
Command-specific subcommands
estat subcommand1 [, options1 ]
estat subcommand2 [, options2 ]
...
estat_summ_options description
-------------------------------------------------------------------------
equation display summary by equation
labels display variable labels
noheader suppress the header
noweights ignore weights
display_options control spacing and display of omitted variables
and base and empty cells
-------------------------------------------------------------------------
eqlist is rarely used and specifies the variables, with optional equation
name, to be summarized. eqlist may be varlist or (eqname1: varlist)
(eqname2: varlist) .... varlist may contain time-series operators; see
tsvarlist.
estat_vce_options description
-------------------------------------------------------------------------
covariance display as covariance matrix; the default
correlation display as correlation matrix
equation(spec) display only specified equations
block display submatrices by equation
diag display submatrices by equation; diagonal blocks
only
format(%fmt) display format for covariances and correlations
nolines suppress lines between equations
display_options control display of omitted variables and base and
empty cells
-------------------------------------------------------------------------
Menu
Statistics > Postestimation > Reports and statistics
Description
estat displays scalar- and matrix-valued statistics after estimation; it
complements predict, which calculates variables after estimation.
Exactly what statistics estat can calculate depends on the previous
estimation command.
Three sets of statistics are so commonly used that they are available
after all estimation commands that store the model log-likelihood. estat
ic displays the Akaike's and Schwarz's Bayesian information criteria.
estat summarize summarizes the variables used by the command and
automatically restricts the sample to e(sample); it also summarizes the
weight variable and cluster structure, if specified. estat vce displays
the covariance or correlation matrix of the parameter estimates of the
previous model.
Option for estat ic
n(#) specifies the N to be used in calculating BIC; see [R] BIC note.
Options for estat summarize
equation requests that the dependent variables and the independent
variables in the equations be displayed in the equation-style format
of estimation commands, repeating the summary information about
variables entered in more than one equation.
labels displays variable labels.
noheader suppresses the header.
noweights ignores the weights, if any, from the previous estimation
command. The default when weights are present is to perform a
weighted summarize on all variables except the weight variable
itself. An unweighted summarize is performed on the weight variable.
display_options: noomitted, vsquish, noemptycells, baselevels,
allbaselevels; see [R] estimation options.
Options for estat vce
covariance displays the matrix as a variance-covariance matrix; this is
the default.
correlation displays the matrix as a correlation matrix rather than a
variance-covariance matrix. rho is a synonym.
equation(spec) selects part of the VCE to be displayed. If spec is
eqlist, the VCE for the listed equations is displayed. If spec is
eqlist1 \ eqlist2, the part of the VCE associated with the equations
in eqlist1 (rowwise) and eqlist2 (columnwise) is displayed. If spec
is *, all equations are displayed. equation() implies block if diag
is not specified.
block displays the submatrices pertaining to distinct equations
separately.
diag displays the diagonal submatrices pertaining to distinct equations
separately.
format(%fmt) specifies the number format for displaying the elements of
the matrix. The default is format(%10.0g) for covariances and
format(%8.4f) for correlations. See [D] format for more information.
nolines suppresses lines between equations.
display_options: noomitted, noemptycells, baselevels, allbaselevels; see
[R] estimation options.
Examples
---------------------------------------------------------------------------
Setup
. sysuse auto
. regress price headroom trunk length mpg
Obtain AIC and BIC
. estat ic
Obtain summary of estimation sample
. estat summarize
---------------------------------------------------------------------------
Setup
. webuse klein
. reg3 (consump wagep wageg) (wagep consump govt capital)
Obtain summary of estimation sample for each equation
. estat summarize, equation
Display VCE for each equation separately, using %7.2f format
. estat vce, block format(%7.2f)
---------------------------------------------------------------------------
Saved results
estat ic saves the following in r():
Matrices
r(S) 1 x 6 matrix of results:
1. sample size 4. degrees of freedom
2. log likelihood of null model 5. AIC
3. log likelihood of full model 6. BIC
estat summarize saves the following in r():
Matrices
r(stats) k x 4 matrix of means, standard deviations, minimums,
and maximums
estat vce saves the following in r():
Matrices
r(V) VCE or correlation matrix
Also see
Manual: [R] estat
Help: [R] estimates, [P] estat programming, [R] summarize
