The robustness of point optimal testing for rosenberg random regression coefficients

Research output: Contribution to journalArticleResearchpeer-review

5 Citations (Scopus)


The literature on testing for the presence of Rosenberg’s (1973) return to normalcy random coefficient model is well developed with both Shively (1988) and Brooks (1993) advocating the use of point optimal tests. This paper explores the robustness of point optimal testing for the Rosenberg alternative to two departures: the special case Hildreth-llouck (1968) alternative and non-normality in regression disturbances, finding the point optimal testing approach to be fairly robust to both departures.

Original languageEnglish
Pages (from-to)35-53
Number of pages19
JournalEconometric Reviews
Issue number1
Publication statusPublished - 1 Jan 1995
Externally publishedYes


  • robustness non-normality
  • varying coefficient regression models point optimal testing

Cite this