st: actest now available from SSC

[Christopher Baum <kit.baum@bc.edu>]

<>
We announce the availability of -actest-, a Stata routine to perform the Cumby-Huizinga general test for autocorrelation in time series

actest performs the general specification test of serial correlation in a time series proposed by Cumby and Huizinga
(C-H, 1990, 1992).  It can be applied to a univariate time series or as a postestimation command after OLS or
instrumental variables (IV) estimation.  The null hypothesis of the test is that the time series is a moving average of
known order q, which could be zero or a positive value. The test considers the general alternative that autocorrelations
of the time series are nonzero at lags greater than q.  The test is general enough to test the hypothesis that the time
series has no serial correlation (q=0) or the null hypothesis that serial correlation in the time series exists, but
dies out at a known finite lag (q>0).

This routine is an enhanced version of our earlier -ivactest-, part of the -ivreg2- package described in Baum, Schaffer, Stillman, 
Stata Journal 2003, 2007. It may also be applied in a panel context.  You may install the routine via

ssc install actest

If you receive an error message relating to the included Mata library, use the replace option on the SSC command.

A presentation discussing its features and relation to other commonly used tests for autocorrelation is available at

http://repec.org/norl13/baum.pdf

An updated version of the presentation will be given at the upcoming UK Stata Users Group meetings in September.

KIt Baum & Mark Schaffer


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/