{smcl}
{* 02jun2003}{...}
{hline}
help for {hi:xtserial}{right:(SJ3-2: st0039)}
{hline}

{title:Wooldridge test for serial correlation in panel-data models}

{p 8 16 2}{cmd:xtserial}
{it:depvar} [{it:varlist}]
{bind:[{cmd:if} {it:exp}]}
{bind:[{cmd:in} {it:range}]}
[{cmd:,}
{cmdab:o:utput}] 

{p 4 4 2}
You must {cmd:tsset} your data before using {cmd:xtserial}; see
help {help tsset}.


{title:Description}

{p 4 4 2}
{cmd:xtserial} implements a test for serial correlation in the idiosyncratic
errors of a linear panel-data model discussed by Wooldridge (2002).  Drukker
(2003) presents simulation evidence that this test has good size and power
properties in reasonable sample sizes.

{p 4 4 2}
Under the null of no serial the residuals from the regression of the
first-differenced variables should have an autocorrelation of -.5.  This
implies that the coefficient on the lagged residuals in a regression of the
lagged residuals on the current residuals should be -.5.  {cmd:xtserial}
performs a Wald test of this hypothesis.  See Drukker (2003) and Wooldridge
(2002) for further details.


{title:Options}

{p 4 8 2}{cmd:output} specifies that the output from the first-differenced
regression should be displayed.  By default, the first-differenced regression
output is not displayed.


{title:Examples}

{p 4 8 2}{cmd:. xtserial ln_wage age* ttl_exp tenure* south, output}


{title:References}

{p 4 8 2}Drukker, D. M. 2003.
Testing for serial correlation in linear panel-data models.
{it:Stata Journal} (3)2: 168-177.

{p 4 8 2}Wooldridge, J. M.  2002.
{it:Econometric Analysis of Cross Section and Panel Data}.
Cambridge, MA: MIT Press.