help ivtobit dialogs: ivtobit svy: ivtobit
also see: ivtobit postestimation
-------------------------------------------------------------------------------
Title
[R] ivtobit -- Tobit model with continuous endogenous regressors
Syntax
Maximum likelihood estimator
ivtobit depvar [varlist1] (varlist2 = varlist_iv) [if] [in] [weight],
ll[(#)] ul[(#)] [mle_options]
Two-step estimator
ivtobit depvar [varlist1] (varlist2 = varlist_iv) [if] [in] [weight],
twostep ll[(#)] ul[(#)] [tse_options]
mle_options description
-------------------------------------------------------------------------
Model
* ll[(#)] lower limit for left censoring
* ul[(#)] upper limit for right censoring
mle use conditional maximum-likelihood estimator;
the default
constraints(constraints) apply specified linear constraints
SE/Robust
vce(vcetype) vcetype may be oim, robust, cluster clustvar,
opg, bootstrap, or jackknife
Reporting
level(#) set confidence level; default is level(95)
first report first-stage estimates
nocnsreport do not display constraints
display_options control spacing and display of omitted
variables and base and empty cells
Maximization
maximize_options control the maximization process
+ coeflegend display coefficients' legend instead of
coefficient table
-------------------------------------------------------------------------
* You must specify at least one of ll[(#)] and ul[(#)].
+ coeflegend does not appear in the dialog box.
tse_options description
-------------------------------------------------------------------------
Model
* twostep use Newey's two-step estimator; the default
is mle
* ll[(#)] lower limit for left censoring
* ul[(#)] upper limit for right censoring
SE/Robust
vce(vcetype) vcetype may be twostep, bootstrap, or
jackknife
Reporting
level(#) set confidence level; default is level(95)
first report first-stage estimates
display_options control spacing and display of omitted
variables and base and empty cells
+ coeflegend display coefficients' legend instead of
coefficient table
-------------------------------------------------------------------------
* twostep is required. You must specify at least one of ll[(#)] and
ul[(#)].
+ coeflegend does not appear in the dialog box.
varlist1 and varlist_iv may contain factor variables; see fvvarlist.
depvar, varlist1, varlist2, and varlist_iv may contain time-series
operators; see tsvarlist.
bootstrap, by, jackknife, rolling, statsby, and svy are allowed; see
prefix.
Weights are not allowed with the bootstrap prefix.
vce(), first, twostep, and weights are not allowed with the svy prefix.
fweights, iweights, and pweights are allowed with the maximum likelihood
estimator. fweights are allowed with Newey's two-step estimator. See
weight.
See [R] ivtobit postestimation for features available after estimation.
Menu
Statistics > Endogenous covariates > Tobit model with endogenous
covariates
Description
ivtobit fits tobit models where one or more of the regressors is
endogenously determined. By default, ivtobit uses maximum-likelihood
estimation. Alternatively, Newey's minimum chi-squared estimator can be
invoked with the twostep option. Both estimators assume that the
endogenous regressors are continuous and so are not appropriate for use
with discrete endogenous regressors. See [R] ivprobit for probit
estimation with endogenous regressors and [R] tobit for tobit estimation
when the model contains no endogenous regressors.
Options for ML estimator
+-------+
----+ Model +------------------------------------------------------------
ll(#) and ul(#) indicate the lower and upper limits for censoring,
respectively. You may specify one or both. Observations with depvar
< ll() are left-censored; observations with depvar > ul() are
right-censored; and remaining observations are not censored. You do
not have to specify the censoring values at all. It is enough to
type ll, ul, or both. When you do not specify a censoring value,
ivtobit assumes that the lower limit is the minimum observed in the
data (if ll is specified) and that the upper limit is the maximum (if
ul is specified).
mle requests that the conditional maximum-likelihood estimator be used.
This is the default.
constraints(constraints); see [R] estimation options.
+-----------+
----+ SE/Robust +--------------------------------------------------------
vce(vcetype) specifies the type of standard error reported, which
includes types that are derived from asymptotic theory, that are
robust to some kinds of misspecification, that allow for intragroup
correlation, and that use bootstrap or jackknife methods; see [R]
vce_option.
+-----------+
----+ Reporting +--------------------------------------------------------
level(#); see [R] estimation options.
first requests that the parameters for the reduced-form equations showing
the relationships between the endogenous variables and instruments be
displayed. For the two-step estimator, first shows the first-stage
regressions. For the maximum likelihood estimator, these parameters
are estimated jointly with the parameters of the tobit equation. The
default is to not show these parameter estimates.
nocnsreport; see [R] estimation options.
display_options: noomitted, vsquish, noemptycells, baselevels,
allbaselevels; see [R] estimation options.
+--------------+
----+ Maximization +-----------------------------------------------------
maximize_options: difficult, technique(algorithm_spec), iterate(#),
[no]log, trace, gradient, showstep, hessian, showtolerance,
tolerance(#), ltolerance(#), nrtolerance(#), nonrtolerance,
from(init_specs); see [R] maximize. This model's likelihood function
can be difficult to maximize, especially with multiple endogenous
variables. The difficult and technique(bfgs) options may be helpful
in achieving convergence.
Setting the optimization type to technique(bhhh) resets the default
vcetype to vce(opg).
The following option is available with ivtobit but is not shown in the
dialog box:
coeflegend; see [R] estimation options.
Options for two-step estimator
+-------+
----+ Model +------------------------------------------------------------
twostep is required and requests that Newey's efficient two-step
estimator be used to obtain the coefficient estimates.
ll(#) and ul(#) indicate the lower and upper limits for censoring,
respectively. You may specify one or both. Observations with depvar
< ll() are left-censored; observations with depvar > ul() are
right-censored; and remaining observations are not censored. You do
not have to specify the censoring values at all. It is enough to
type ll, ul, or both. When you do not specify a censoring value,
ivtobit assumes that the lower limit is the minimum observed in the
data (if ll is specified) and that the upper limit is the maximum (if
ul is specified).
+-----------+
----+ SE/Robust +--------------------------------------------------------
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 [R] vce_option.
+-----------+
----+ Reporting +--------------------------------------------------------
level(#); see [R] estimation options.
first requests that the parameters for the reduced-form equations showing
the relationship between the endogenous variables and instruments be
displayed. For the two-step estimator, first shows the first-stage
regressions. For the maximum likelihood estimator, these parameters
are estimated jointly with the parameters of the tobit equation. The
default is to not show these parameter estimates.
display_options: noomitted, vsquish, noemptycells, baselevels,
allbaselevels; see [R] estimation options.
The following option is available with ivtobit but is not shown in the
dialog box:
coeflegend; see [R] estimation options.
Examples
Setup
. webuse laborsup
Obtain full ML estimates
. ivtobit fem_inc fem_educ kids (other_inc = male_educ), ll
. ivtobit fem_inc fem_educ kids (other_inc = male_educ), ll(12)
Obtain two-step estimates
. ivtobit fem_inc fem_educ kids (other_inc = male_educ), ll twostep
. ivtobit fem_inc fem_educ kids (other_inc = male_educ), ll(12)
twostep
Saved results
ivtobit, mle saves the following in e():
Scalars
e(N) number of observations
e(N_unc) number of uncensored observations
e(N_lc) number of left-censored observations
e(N_rc) number of right-censored observations
e(llopt) contents of ll()
e(ulopt) contents of ul()
e(k) number of parameters
e(k_eq) number of equations
e(k_eq_model) number of equations in model Wald test
e(k_aux) number of auxiliary parameters
e(k_dv) number of dependent variables
e(k_autoCns) number of base, empty, and omitted constraints
e(df_m) model degrees of freedom
e(ll) log likelihood
e(N_clust) number of clusters
e(k_eq_skip) identifies which equations should not be reported
in the coefficient table
e(endog_ct) number of endogenous regressors
e(p) model Wald p-value
e(p_exog) exogeneity test Wald p-value
e(chi2) model Wald chi-squared
e(chi2_exog) Wald chi-squared test of exogeneity
e(rank) rank of e(V)
e(ic) number of iterations
e(rc) return code
e(converged) 1 if converged, 0 otherwise
Macros
e(cmd) ivtobit
e(cmdline) command as typed
e(depvar) name of dependent variable
e(instd) instrumented variables
e(insts) instruments
e(wtype) weight type
e(wexp) weight expression
e(title) title in estimation output
e(clustvar) name of cluster variable
e(chi2type) Wald; type of model chi-squared test
e(vce) vcetype specified in vce()
e(vcetype) title used to label Std. Err.
e(diparm#) display transformed parameter #
e(method) ml
e(opt) type of optimization
e(which) max or min; whether optimizer is to perform
maximization or minimization
e(ml_method) type of ml method
e(user) name of likelihood-evaluator program
e(technique) maximization technique
e(singularHmethod) m-marquardt or hybrid; method used when Hessian is
singular
e(crittype) optimization criterion
e(properties) b V
e(predict) program used to implement predict
e(footnote) program used to implement the footnote display
e(marginsok) predictions allowed by margins
e(asbalanced) factor variables fvset as asbalanced
e(asobserved) factor variables fvset as asobserved
Matrices
e(b) coefficient vector
e(Cns) constraints matrix
e(ilog) iteration log (up to 20 iterations)
e(gradient) gradient vector
e(Sigma) Sigma hat
e(V) variance-covariance matrix of the estimators
e(V_modelbased) model-based variance
e(ml_h) derivative tolerance, (abs(b)+1e-3)*1e-3
e(ml_scale) derivative scale factor
Functions
e(sample) marks estimation sample
ivtobit, twostep saves the following in e():
Scalars
e(N) number of observations
e(N_unc) number of uncensored observations
e(N_lc) number of left-censored observations
e(N_rc) number of right-censored observations
e(llopt) contents of ll()
e(ulopt) contents of ul()
e(df_m) model degrees of freedom
e(df_exog) degrees of freedom for chi-squared test of
exogeneity
e(p) model Wald p-value
e(p_exog) exogeneity test Wald p-value
e(chi2) model Wald chi-squared
e(chi2_exog) Wald chi-squared test of exogeneity
e(rank) rank of e(V)
Macros
e(cmd) ivtobit
e(cmdline) command as typed
e(depvar) name of dependent variable
e(instd) instrumented variables
e(insts) instruments
e(wtype) weight type
e(wexp) weight expression
e(chi2type) Wald; type of model chi-squared test
e(vce) vcetype specified in vce()
e(vcetype) title used to label Std. Err.
e(method) twostep
e(properties) b V
e(predict) program used to implement predict
e(footnote) program used to implement the footnote display
e(marginsok) predictions allowed by margins
e(asbalanced) factor variables fvset as asbalanced
e(asobserved) factor variables fvset as asobserved
Matrices
e(b) coefficient vector
e(V) variance-covariance matrix of the estimators
e(V_modelbased) model-based variance
Functions
e(sample) marks estimation sample
Also see
Manual: [R] ivtobit
Help: [R] ivtobit postestimation;
[R] gmm, [R] ivprobit, [R] ivregress, [R] regress, [SVY] svy
estimation, [R] tobit, [XT] xtintreg, [XT] xttobit
