help xthtaylor dialog: xthtaylor
also see: xthtaylor postestimation
-------------------------------------------------------------------------------
Title
[XT] xthtaylor -- Hausman-Taylor estimator for error-components models
Syntax
xthtaylor depvar indepvars [if] [in] [weight] , endog(varlist)
[options]
options description
-------------------------------------------------------------------------
Main
noconstant suppress constant term
* endog(varlist) explanatory variables in indepvars to be treated
as endogenous
constant(varlist_ti) independent variables that are constant within
panel
varying(varlist_tv) independent variables that are time varying
within panel
amacurdy fit model based on Amemiya and MaCurdy estimator
SE
vce(vcetype) vcetype may be conventional, bootstrap, or
jackknife
Reporting
level(#) set confidence level; default is level(95)
small report small-sample statistics
-------------------------------------------------------------------------
* endog(varlist) is required.
A panel variable must be specified. For xthtaylor, amacurdy, a time
variable must also be specified. Use xtset.
depvar, indepvars, and all varlists may contain time-series operators;
see tsvarlist.
by, statsby, and xi are allowed; see prefix.
iweights and fweights are allowed unless the amacurdy option is
specified. Weights must be constant within panel; see weight.
See [XT] xthtaylor postestimation for features available after
estimation.
Menu
Statistics > Longitudinal/panel data > Endogenous covariates >
Hausman-Taylor regression (RE)
Description
xthtaylor fits panel-data random-effects models in which some of the
covariates are correlated with the unobserved individual-level random
effect. The estimators, originally proposed by Hausman and Taylor (1981)
and Amemiya and MaCurdy (1986), are based on instrumental variables. By
default, xthtaylor uses the Hausman-Taylor estimator. When the amacurdy
option is specified, xthtaylor uses the Amemiya-MaCurdy estimator.
Although the estimators implemented in xthtaylor and xtivreg (see [XT]
xtivreg) use the method of instrumental variables, each command is
designed for different problems. The estimators implemented in xtivreg
assume that a subset of the explanatory variables in the model are
correlated with the idiosyncratic error e[i,t]. In contrast, the
Hausman-Taylor and Amemiya-MaCurdy estimators that are implemented in
xthtaylor assume that some of the explanatory variables are correlated
with the individual-level random effects, u[i], but that none of the
explanatory variables are correlated with the idiosyncratic error e[i,t].
Options
+------+
----+ Main +-------------------------------------------------------------
noconstant; see [R] estimation options.
endog(varlist) specifies that a subset of explanatory variables in
indepvars be treated as endogenous variables, i.e., the explanatory
variables that are assumed to be correlated with the unobserved
random effect. endog() is required.
constant(varlist_ti) specifies the subset of variables in indepvars that
are time invariant, that is, constant within panel. By using this
option, you assert not only that the variables specified in
varlist_ti are time invariant, but also that all other variables in
indepvars are time varying. If this assertion is false, xthtaylor
does not perform the estimation and will issue an error message.
xthtaylor automatically detects which variables are time invariant
and which are not. However, users may want to check their
understanding of the data and specify which variables are time
invariant and which are not.
varying(varlist_tv) specifies the subset of the variables in indepvars
that are time varying. By using this option, you assert not only
that the variables specified in varlist_tv are time varying, but also
that all other variables in indepvars are time-invariant. If this
assertion is false, xthtaylor does not perform the estimation and
will issue an error message. xthtaylor automatically detects which
variables are time varying and which are not. However, users may
want to check their understanding of the data and specify which
variables are time varying and which are not.
amacurdy specifies that the Amemiya-MaCurdy estimator be used. This
estimator uses extra instruments to gain efficiency at the cost of
additional assumptions on the data-generating process. This option
may be specified only for samples containing balanced panels, and
weights may not be specified. The panels must also have a common
initial period.
+----+
----+ SE +---------------------------------------------------------------
vce(vcetype) specifies the type of standard error reported, which
includes types that are derived from asymptotic theory and that use
bootstrap or jackknife methods; see [XT] vce_options.
vce(conventional), the default, uses the conventionally derived
variance estimator for this Hausman-Taylor model.
+-----------+
----+ Reporting +--------------------------------------------------------
level(#); see [R] estimation options.
small specifies that the p-values from the Wald tests in the output and
all subsequent Wald tests obtained via test use t and F distributions
instead of the large-sample normal and chi-squared distributions. By
default, the p-values are obtained using the normal and chi-squared
distributions.
Examples
Setup
. webuse psidextract
Hausman-Taylor estimates
. xthtaylor lwage wks south smsa ms exp exp2 occ ind union fem blk
ed, endog(exp exp2 occ ind union ed) constant(fem blk ed)
Amemiya-MaCurdy estimates
. xthtaylor lwage wks south smsa ms exp exp2 occ ind union fem blk
ed, endog(exp exp2 occ ind union ed) amacurdy
Saved results
xthtaylor saves the following in e():
Scalars
e(N) number of observations
e(N_g) number of groups
e(df_m) model degrees of freedom
e(df_r) residual degrees of freedom (small only)
e(g_min) smallest group size
e(g_avg) average group size
e(g_max) largest group size
e(Tcon) 1 if panels balanced; 0 otherwise
e(sigma_u) panel-level standard deviation
e(sigma_e) standard deviation of epsilon_it
e(chi2) chi-squared
e(rho) rho
e(F) model F (small only)
e(Tbar) harmonic mean of group sized
e(rank) rank of e(V)
Macros
e(cmd) xthtaylor
e(cmdline) command as typed
e(depvar) name of dependent variable
e(ivar) variable denoting groups
e(tvar) time variable, amacurdy only
e(TVexogenous) exogenous time-varying variables
e(TIexogenous) exogenout time-invariant variables
e(TVendogenous) endogenous time-varying variables
e(TIendogenous) endogenous time-invariant variables
e(wtype) weight type
e(wexp) weight expression
e(title) Hausman-Taylor or Amemiya-MaCurdy
e(chi2type) Wald; type of model chi-squared test
e(vce) vcetype specified in vce()
e(vcetype) title used to label Std. Err.
e(properties) b V
e(predict) program used to implement predict
Matrices
e(b) coefficient vector
e(V) variance-covariance matrix of the estimators
Functions
e(sample) marks estimation sample
Also see
Manual: [XT] xthtaylor
Help: [XT] xthtaylor postestimation;
[XT] xtset; [XT] xtivreg, [XT] xtreg
