{smcl}
{* 19May06}{...}
{hline}
help for {hi:xtcsd}{right:(SJ6-4: st0113)}
{hline}

{title:Test for cross-sectional dependence in panel-data models}

{p 8 16 2}
{cmd: xtcsd} {cmd:,} [{cmdab:pes:aran} {cmdab:fri:edman} {cmdab:fre:es}
                      {cmdab:abs} {cmdab:show}]  


{p 4 8 2} {cmd:xtcsd} is a postestimation command for use with cross-section
time-series data following {cmd:fe} or {cmd:re} models.  You must
{helpb tsset} your data before using {cmd:xtcsd}.


{title:Description}

{p 4 4 2} {cmd:xtcsd} tests for cross-sectional dependence in fixed-effects or
random-effects panel-data models.  One of the command's options, {cmd:pesaran},
{cmd:friedman}, or {cmd:frees}, must be specified. 
	
{p 4 4 2} A standard assumption in panel-data models ({helpb xtreg}) is that
the error terms are independent across cross sections.  This assumption is
used for identification purposes rather than descriptive accuracy. In the
context of large T and small N, the Lagrange multiplier test statistic
proposed by Breusch and Pagan (1980) can be used to test for cross-sectional
dependence (see {helpb xttest2}).  However, cross-sectional
time-series datasets usually come in the form of small T and large N. In this
case, the Breusch-Pagan test is not valid. 

{p 4 4 2} {cmd:xtcsd} test the hypothesis of cross-sectional independence in
panel-data models with small T and large N by implementing two semiparametric
tests proposed by Friedman (1937) and Frees (1995, 2004), as well as the
parametric testing procedure proposed by Pesaran (2004).  {cmd:xtcsd} with
option {cmd:pesaran} can handle balanced as well as unbalanced panels.   


{title:Options}

{p 4 8 2} {cmd:pesaran} tests for cross-sectional dependence following the
methods shown in Pesaran (2004). Pesaran's statistic follows a standard normal
distribution and can handle balanced and unbalanced panels.  
			 
{p 4 8 2} {cmd:friedman} tests for cross-sectional dependence by using
Friedman's chi-squared distributed statistic.  For unbalanced panels,
Friedman's test uses only the observations available for all cross-sectional
units.

{p 4 8 2} {cmd:frees} tests for cross-sectional dependence by using Frees' Q
distribution (T-asymptotically distributed).  For unbalanced panels, Frees'
test uses only the observations available for all cross-sectional units.    
	
{p 4 8 2} {cmd:abs} computes the average absolute value of the off-diagonal
elements of the cross-sectional correlation matrix of residuals.

{p 4 8 2} {cmd:show} shows the cross-sectional correlation matrix of
residuals. 
	

{title:Notes}

{p 4 8 2} The small-sample comparative performance of these tests under
various model specifications is examined in Sarafidis and De Hoyos (2006). 


{title:Examples}

{p 4 8 2}{cmd:. use "http://www.econ.cam.ac.uk/phd/red29/xtcsd_baltagi.dta"}
{p_end}
{p 4 8 2}{cmd:. xtreg  lngsp lnpcap lnpc lnemp unemp, fe}{p_end}
{p 4 8 2}{cmd:. xtcsd, frees}{p_end}
{p 4 8 2}{cmd:. xtcsd, pesaran show}{p_end}
{p 4 8 2}{cmd:. xtreg  lngsp lnpcap lnpc lnemp unemp, re}{p_end}
{p 4 8 2}{cmd:. xtcsd, friedman show abs}{p_end}


{title:References}

{p 4 8 2}Frees, E. W. 1995. Assessing cross-sectional correlations in panel
data. {it:Journal of Econometrics} 69: 393-414.

{p 4 8 2}------. 2004. {it:Longitudinal and Panel Data: Analysis and}
{it:Applications in the Social Sciences}. Cambridge: Cambridge University
Press. 

{p 4 8 2}Friedman, M. 1937. The use of ranks to avoid the assumption of
normality implicit in the analysis of variance.
{it:Journal of the American Statistical Association} 32: 675-701.

{p 4 8 2}Pesaran, M. H. 2004. General diagnostic tests for cross section
dependence in panels. University of Cambridge, Faculty of Economics, Cambridge
Working Papers in Economics No. 0435.

{p 4 8 2}Sarafidis, V., and R. E. De Hoyos. 2006. On testing for cross
sectional dependence in panel data models. Mimeo, University of
Cambridge.


{title:Acknowledgments}

{p 4 8 2} Our code benefited greatly from Christopher F. Baum's
{helpb xttest2}. We thank David Drukker for useful suggestions. 


{title:Authors}

    Rafael E. De Hoyos
    Development Prospects Group, The World Bank
    {browse "mailto:rdehoyos@worldbank.org":rdehoyos@worldbank.org} 

    {browse "http://www.econ.usyd.edu.au/15583.html":Vasilis Sarafidis}
    Discipline of Econometrics and Business Statistics, University of Sydney
    {browse "mailto:V.Sarafidis@econ.usyd.edu.au":V.Sarafidis@econ.usyd.edu.au}


{title:Also see}

{psee}Manual:	{hi:[U] 23 Estimation and postestimation commands}{break}
	{hi:[XT] xtreg}

{psee}Online:	{helpb xttest2} (if installed), {helpb bpagan} (if installed),
          {helpb ivreg2} (if installed)
{p_end}