Recursive estimation and generated regressors

Michael McAleer, C. R. McKenzie

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This paper considers the estimation of models containing anticipated and unanticipated variables using two-step methods, in which the unobserved variables are generated by recursive regressions. It is shown that the parameter estimates of the structural equation are inefficient and the conventionally programmed OLS standard errors are inconsistent.

Original languageEnglish
Pages (from-to)1-5
Number of pages5
JournalEconomics Letters
Volume39
Issue number1
DOIs
Publication statusPublished - 1992 May

    Fingerprint

ASJC Scopus subject areas

  • Finance
  • Economics and Econometrics

Cite this