Description
xtdpdserial: Panel data serial correlation tests
xtdpdserial implements serial correlation tests for linear panel data models that fall into the framework of the portmanteau test developed by Jochmans (2020). Special cases are the Arellano and Bond (1991) and Yamagata (2008) tests, and further variants proposed by Kripfganz, Demetrescu, and Hosseinkouchack (2024) designed to improve the power of the test. By default, these are postestimation tests for serial correlation in the idiosyncratic error component from a linear error components model. The test accounts for the estimation error in the regression residuals and is robust to heteroskedasticity. Alternatively, instead of using residuals, the command can be used to test for serial correlation in a specified variable.
Installation & Help
To install a fresh version of this package, type the following in Stata's command window:
→ net install xtdpdserial, from(http://www.kripfganz.de/stata/)
or (less frequently updated):
→ ssc install xtdpdserial
To display the currently installed version of this package, type the following in Stata's command window:
→ which xtdpdserial
To update an existing version of this package, type the following in Stata's command window:
→ adoupdate xtdpdserial, update
To access the help files of this package after the installation, type the following in Stata's command window:
→ help xtdpdserial
Accompanying Article
