Finite-sample power of tests for autocorrelation in models containing lagged dependent variables

B. A. Inder

Research output: Contribution to journalArticleResearchpeer-review

18 Citations (Scopus)


Monte Carlo evidence is presented to indicate that for a given size of the test, the Durbin-Watson test is more powerful than Durbin's h test in testing for autocorrelation in models with a lagged dependent variable.

Original languageEnglish
Pages (from-to)179-185
Number of pages7
JournalEconomics Letters
Issue number2-3
Publication statusPublished - 1 Jan 1984

Cite this