help bootstrap postestimation dialog: estat
also see: bootstrap
-------------------------------------------------------------------------------
Title
[R] bootstrap postestimation -- Postestimation tools for bootstrap
Description
The following postestimation command is of special interest after
bootstrap:
command description
-------------------------------------------------------------------------
estat bootstrap percentile-based and bias-corrected CI tables
-------------------------------------------------------------------------
The following standard postestimation commands are also available:
command description
-------------------------------------------------------------------------
estat AIC, BIC, VCE, and estimation sample summary
estimates cataloging estimation results
* hausman Hausman's specification test
* lincom point estimates, standard errors, testing, and
inference for linear combinations of coefficients
* margins marginal means, predictive margins, marginal
effects, and average marginal effects
* nlcom point estimates, standard errors, testing and
inference for nonlinear combinations of
coefficients
* predict predictions, residuals, influence statistics, and
other diagnostic measures
* predictnl point estimates, standard errors, testing, and
inference for generalized predictions
* test Wald tests of simple and composite linear hypotheses
* testnl Wald tests of nonlinear hypotheses
-------------------------------------------------------------------------
* This postestimation command is allowed if it may be used after command.
Special-interest postestimation command
estat bootstrap displays a table of confidence intervals for each
statistic from a bootstrap analysis.
Syntax for predict
The syntax of predict (and even if predict is allowed) following
bootstrap depends upon the command used with bootstrap. If predict is
not allowed, neither is predictnl.
Syntax for estat bootstrap
estat bootstrap [, options]
options description
-------------------------------------------------------------------------
bc bias-corrected CIs; the default
bca bias-corrected and accelerated (BC_a) CIs
normal normal-based CIs
percentile percentile CIs
all all available CIs
noheader suppress table header
nolegend suppress table legend
verbose display the full table legend
-------------------------------------------------------------------------
bc, bca, normal, and percentile may be used together.
Menu
Statistics > Postestimation > Reports and statistics
Options for estat bootstrap
bc is the default and displays bias-corrected confidence intervals.
bca displays bias-corrected and accelerated confidence intervals. This
option assumes that you also specified the bca option on the
bootstrap prefix command.
normal displays normal approximation confidence intervals.
percentile displays percentile confidence intervals.
all displays all available confidence intervals.
noheader suppresses display of the table header. This option implies
nolegend.
nolegend suppresses display of the table legend, which identifies the
rows of the table with the expressions they represent.
verbose requests that the full table legend be displayed.
Examples
Setup
. sysuse auto
. bootstrap, reps(100) bca: regress mpg weight gear foreign
Obtain bias-corrected CIs
. estat bootstrap
Obtain bias-corrected and accelerated CIs
. estat bootstrap, bca
Obtain all available CIs
. estat bootstrap, all
Test that coefficients on gear and foreign sum to 0
. test gear + foreign = 0
Compute estimate, SE, test statistic, significance level, and CI for the
ratio of coefficient of gear_ratio to the coefficient of foreign
. nlcom _b[gear_ratio] / _b[foreign]
Also see
Manual: [R] bootstrap postestimation
Help: [R] bootstrap
