{smcl}
{* 01dec2008}{...}
help for {cmd:test_condate}
{hline}

{title:Command test_condate}

{pstd}
Tests Zero Conditional ATE, Constant Conditional ATE, and Zero ATE hypotheses as in Crump, R. K.,V. J. Hotz, G. W. Imbens 
and O. A. Mitnik (2008) "Nonparametric Tests for Treatment Effect Heterogeneity", Review of Economics and Statistics, 
August, 90(3): 389-405.


{title:Syntax}

{p 8 17 2}
{cmd:test_condate}
{depvar}
{varlist}
{ifin} {weight}{cmd:, tvar({varname})}
[{cmdab:nor:obust}]

{synoptset 20 tabbed}{...}
{synopthdr}
{synoptline}
{syntab:Main}
{synopt:{opth tvar(varname)}}Treatment indicator variable{p_end}
{synopt:{opt nor:obust}}Uses homoskedastic var-cov matrix (default: heteroskedastic-robust){p_end}
{synoptline}
{syntab:{cmd:weight}s are allowed; see {help weight}.}
{p2colreset}{...}



{title:Description}

{pstd}
{cmd:test_condate} Tests zero conditional average treatment effects, constant conditional average treatment effects, 
and zero average treatment effects hypotheses
as in Crump, Richard K.,V. Joseph Hotz, Guido W. Imbens and Oscar A. Mitnik (2008) 
"Nonparametric Tests for Treatment Effect Heterogeneity",
Review of Economics and Statistics, August, 90(3): 389-405. 

{pstd}
When estimating the regression functions for 
treated and control groups, the command makes sure that the same variables can be used for both regression functions. 
Those variables that cannot be used for either group are dropped from the estimation. A macro in the r() results 
(see below) collects the names of the variables actually used in the estimation.


{title:Options}

{dlgtab:Main}

{phang}
{opth tvar(varname)} Treatment indicator variable (1=treatment group, 0=control group).

{phang}
{opt nor:obust} Uses homoskedastic var-cov matrix. The default is to use a heteroskedastic-robust var-cov matrix.


{title:Examples}

{phang}{cmd:. test_condate y xvar1 xvar2 xvar3, tvar(treat)}

{phang}{cmd:. test_condate y xvar1 xvar2 xvar3, tvar(treat) norobust}


{title:Saved results}

{pstd}
{cmd:test_condate} saves the following in {cmd:r()}:

{synoptset 20 tabbed}{...}
{p2col 5 20 24 2: Macros}{p_end}
{synopt:{cmd:r(variables_used)}}list of variables actually used in the estimation of the regression functions{p_end}
{synopt:{cmd:r(tests_order)}}indicates which test is in each column of the matrices below{p_end}

{p2col 5 20 24 2: Matrices}{p_end}
{synopt:{cmd:r(Qtests)}}chi-squared version of the tests (Q statistics){p_end}
{synopt:{cmd:r(dfQtests)}}dof of the chi-squared tests{p_end}
{synopt:{cmd:r(pQtests)}}p-value of the chi-squared tests{p_end}
{synopt:{cmd:r(Ttests)}}normal version of the tests (T statistics){p_end}
{synopt:{cmd:r(pTtests)}}p-value of the normal tests{p_end}
{p2colreset}{...}