{smcl}
{* 9oct2002}{...}
{hline}
help for {hi:atts}{right:(SJ2-4:st0026)}
{hline}

{title:Calculate the average treatment effect on the treated using stratification matching}

{p 12 17 2}
{cmd:atts}
{it:outcome treatment } 
[{cmd:if} {it:exp}]
[{cmd:in} {it:range}]
{cmd:,} 
{cmd:pscore}{cmd:(}{it:scorevar}{cmd:)} 
{cmd:blockid}{cmd:(}{it:blockvar}{cmd:)} 
[
{cmdab:comsup}
{cmdab:det:ail}
{cmdab:boot:strap }
{cmdab:r:eps(}{it:#}{cmd:) }
{cmdab:noi:sily}
{cmdab:d:ots}
]


{title:Description}

{p 4 4 2} {cmd:atts} estimates the average treatment effect on the treated
(ATT) using stratification matching. Users have to provide the name of a
variable specifying the estimated propensity score and the name of a variable
specifying a block identifier of the estimated propensity score; both of which
can be produced with {cmd:pscore}. Since {cmd:atts} is based on the same
stratification procedure used for estimating the propensity score, it can most
naturally be used after running {cmd:pscore} because by construction, in each
block defined by this procedure, the covariates are balanced and the assignment
to treatment can be considered random. 

{p 4 4 2} It is left to the responsibility of the user to select the
{cmd:comsup} option if the user provided propensity score and structure of
blocks requires it. However, note that this option becomes irrelevant for the
computation of the ATT when running {cmd:atts} after using {cmd:pscore} with
the {cmd:comsup} option. In this case, the {cmd:blockid} variable is defined
only on the common support. Hence, the ATT can be computed only on the common
support. However, it is advisable to impose the {cmd:comsup} option in order
to obtain correct descriptive statistics for the number of treated and
controls.

{p 4 4 2} The ATT is computed using a weighted (by the number of treated) 
average of the block-specific treatment effects. In turn, these are 
computed as the difference in average outcomes of treated and controls 
within the same block for which {cmd:pscore} has found all the control 
variables to be balanced.


{title:Options}

{p 4 8 2}  
{cmd:pscore(}{it:scorevar}{cmd:)} is a compulsory option which specifies 
the name of the user-provided variable containing the estimated 
propensity score.

{p 4 8 2}  
{cmd:blockid(}{it:blockvar}{cmd:)} is a compulsory option which specifies 
the name of the user-provided variable containing the block identifier of 
the estimated propensity score.

{p 4 8 2} 
{cmd:comsup} restricts the computation of the ATT to the region of
common support.

{p 4 8 2} 
{cmd:detail} displays more detailed output documenting the steps performed to
obtain the final results.

{p 4 8 2}                                                      
{cmd:bootstrap} bootstraps the standard error of the treatment effect.

{p 4 8 2}{cmd:reps(}{it:#}{cmd:)} specifies the number of bootstrap
replications to be performed. The default is 50.  This option produces an
effect only if the {cmd:bootstrap} option is specified.

{p 4 8 2}{cmd:noisily} requests that any output from the bootstrap
replications be displayed.  This option produces an effect only if the
{cmd:bootstrap} option is specified.

{p 4 8 2}{cmd:dots} requests that a dot be placed on the screen at the
beginning of each bootstrap replication.  This option produces an effect only
if the {cmd:bootstrap} option is specified.


{title:Remarks}

{p 4 4 2} Please remember to use the {cmd:update query} command before running
this program to make sure you have an up-to-date version of Stata installed.
Otherwise, this program may not run properly.

{p 4 4 2} The treatment has to be binary.

{p 4 4 2} The bootstrapping procedure does not re-estimate the propensity score 
at each iteration. This is because different bootstrapped samples imply
different specifications of the propensity score satisfying the balancing
property. Consequently, different observations could end up in different
blocks in different bootstrap iterations. This in turn, would preclude the
computation of the ATT block by block.

{p 4 4 2} Note that there are two special cases as concerns for the
computation of the ATT and its analytical standard error. First, if there is
no treated and/or no control unit in one (or more) of the blocks, the ATT is
computed on the remaining blocks, which practically amounts to imposing a
(block-based) common support condition.  Second, if there is exactly one
treated and/or one control in one (or more) of the blocks, the ATT in that
block can still be computed but the standard error cannot. In this case,
{cmd:atts} will produce missing values for the standard error.  However,
bootstrapped standard errors can still be computed.


{title:Saved results} 

{p 4 4 2} The program stores the estimated treatment
effect, its standard error and the t statistics respectively in the
scalars {cmd:r(atts)}, {cmd:r(seatts)}, and {cmd:r(tsatts)}.

{p 4 4 2} The number of treated and the number of controls are stored 
respectively in the scalars {cmd:r(nts)} and {cmd:r(ncs)}.

{p 4 4 2} The bootstrapped standard error and t statistic are stored 
respectively in the scalars {cmd:r(bseatts)} and {cmd:r(btsatts)}.


{title:Example}

{p 4 8 2}{cmd:. atts wage training, pscore(myscore) blockid(myblock) }

{p 4 8 2}{cmd:. atts wage training, pscore(myscore) blockid(myblock) boot reps(100) dots}

{p 4 8 2}{cmd:. atts wage training, pscore(myscore) blockid(myblock) comsup boot reps(100) dots}



{title:Authors} 
 
	{browse "http://www.sobecker.de":Sascha O. Becker} 
	Center for Economic Studies, University of Munich 
  
	{browse "http://www.iue.it/Personal/Ichino":Andrea Ichino} 
	Department of Economics, European University Institute, Florence 
  
{p 4 4 2}Email {browse "mailto:so.b@gmx.net":so.b@gmx.net} or {browse "mailto:andrea.ichino@iue.it":andrea.ichino@iue.it} if you observe any problems. 
			
	
{title:Also see}

{p 4 14 2}Online: help for {help pscore}, {help attr}, {help attk},
{help attnw}, {help attnd} (if installed), and {help bs}. 

{p 14 14 2}Further details on the analytical formulas and on the algorithms 
used in these programs  can be found under  
{browse "http://www.sobecker.de":http://www.sobecker.de} 
or 
{browse "http://www.iue.it/Personal/Ichino":http://www.iue.it/Personal/Ichino}.