Spanish Economic Review

, Volume 1, Issue 1, pp 79–90 | Cite as

Econometric applications of positive rank-one modifications of the symmetric factorization of a positive semi-definite matrix

  • Enrique Sentana
  • 26 Downloads

Abstract.

We present an algorithm for updating the symmetric factorization of a positive semi-definite matrix after a positive rank-one modification, which works even if the matrices involved do not have full rank. Recursive least squares and factor analysis provide two important econometric applications. An illustrative simulation shows that it can be potentially very useful in recursive situations.

JEL classification: C22, C32 
Key words: Recursive least squares, factor analysis, Cholesky decomposition, multicollinearity 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 1999

Authors and Affiliations

  • Enrique Sentana
    • 1
  1. 1. CEMFI, Casado del Alisal 5, E-28014 Madrid, Spain (e-mail: sentana@cemfi.es) ES

Personalised recommendations