{smcl}
{* 11jan2006}{...}
{hline}
help for {hi:condivreg}
{hline}

{title:Instrumental variables regression with potentially weak instruments}

{p 8 14 2}{cmd:condivreg} {it:depvar} [{it:varlist1}]
{cmd:(}{it:endogvar}{cmd:=}{it:varlist_iv}{cmd:)} 
[{cmd:if} {it:exp}] [{cmd:in} {it:range}]
{cmd:[,} {{cmd:2sls} | {cmd:liml}}
{cmd:nocons} {cmd:noinstcons}  {cmd:test}{cmd:(}{it:#}{cmd:)}
{cmd:ar} {cmd:lm} {cmd:interval} {cmd:level}{cmd:(}{it:#}{cmd:)}
{cmd:]}


{title:Description}

{p 4 4 2}
{cmd:condivreg} fits a linear regression using 2SLS or LIML of
{it:depvar} on {it:varlist1} and {it:endogvar} using {it:varlist_iv}
(along with {it:varlist1}) as instruments for {it:varlist2}.  It also
reports the confidence region and {it:p}-value for the coefficient on
the single endogenous variable based on the conditional likelihood ratio
(CLR) approach developed by Moreira (2003). Computation of the
conditional {it:p}-value for the CLR test uses the algorithm of Andrews,
Moreira and Stock (2005). Andrews, Moreira and Stock (2004) showed that
the CLR test is approximately optimal.  In particular, it dominates the
Anderson and Rubin (1949) test and the Lagrange Multiplier (score) test
proposed independently by Kleibergen (2002) and Moreira (2001). Results
for the Anderson-Rubin and score tests are optionally provided. 


{title:Options}

{p 4 8 2}
{cmd:2sls} requests that the 2SLS estimator be used. {cmd:2sls} is the default.

{p 4 8 2}
{cmd:liml} requests that the LIML estimator be used. {cmd:2sls} and
{cmd:liml} are mutually exclusive.

{p 4 8 2}
{cmd:nocons} indicates that no constant term is to be included in the
regression equation. The default is to include a constant term.

{p 4 8 2}
{cmd:noinstcons} indicates that no constant term is to be included in
the first-stage regression of the endogenous variable on the instruments
and exogenous variables. Stata's {cmd:ivreg} command excludes a constant
from both equations if its {cmd:noconstant} option is specified. 
Usually one will not want to specify {cmd:noinstcons} unless
{cmd:nocons} is also specified, but we give the user the option to
experiment. By default a constant term is included.

{p 4 8 2}
{cmd:test(}{it:#}{cmd:)} contains the hypothesized value of the
endogenous variable's coefficient. The default is {cmd:test(0)}.

{p 4 8 2}
{cmd:ar} provides the coverage-corrected confidence set and
size-corrected test based on the Anderson-Rubin statistic. 

{p 4 8 2}
{cmd:lm} provides the coverage-corrected confidence set and
size-corrected test based on the Lagrange multiplier (score) statistic.

{p 4 8 2}
{cmd:interval} displays the confidence interval, which is the minimal
convex interval containing the coverage-corrected confidence set. 

{p 4 8 2}
{cmd:level(}{it:#}{cmd:)} specifies the nominal significance level, as 
a percentage, to be used when displaying results.  The default is 
{cmd:level(95)} or as set by {cmd:set level}.  See {help level}.


{title:Example}

{p 4 8 2}{cmd:. condivreg y1 x1 (y2 = z1 z2 z3), liml interval}{p_end}

{p 4 8 2}{cmd:. condivreg y1 x1 (y2 = z1 z2 z3 z4), ar lm test(0.1) }{p_end}


{title:Also see}

{p 4 8 2}Anderson, T.W. and H. Rubin (1949): "Estimators of the Parameters of a
Single Equation in a Complete Set of Stochastic Equations,"
{it:Annals of Math. Statistics, 21, 570-582.}

{p 4 8 2}Andrews, D., Moreira, M. and J. Stock (2004): "Optimal Invariant Tests for
Instrumental Variables Regression," {it:Unpublished manuscript.}

{p 4 8 2}Andrews, D., Moreira, M. and J. Stock (2005): "Performance of Conditional 
Wald Tests in IV Regression with Weak Instruments," {it:Journal of Econometrics,}
forthcoming.

{p 4 8 2}Kleibergen, F. (2002): "Pivotal Statistics for Testing Structural Parameters in
Instrumental Variables Regression," {it:Econometrica, 70, 1781-1803.}

{p 4 8 2}Moreira, M. (2001): "Tests with Correct Size when Instruments Can 
Be Arbitrarily Weak," {it:Center for Labor Economics Working Paper, 37,} 
UC Berkeley.

{p 4 8 2}Moreira, M. (2003): "A Conditional Likelihood Ratio Test for
Structural Models," {it:Econometrica, 71, 1027-1048.}

{p 4 13 2}Manual:  {hi:[R] ivreg}
{p 4 13 2}On-line:  help for {help condtest}, {help condgraph}; {help ivreg}