Statistical analysis of curved probability densities

Masanobu Taniguchi, Yoshihide Watanabe

Research output: Contribution to journalArticlepeer-review

7 Citations (Scopus)


Suppose that pn(· ; θ) is the joint probability density of n observations which are not necessarily i.i.d. In this paper we discuss the estimation of an unknown parameter u of a family of "curved probability densities" defined by M = {pn(· ; θ(u)), dim u < dim θ} embedded in S = {pn(· ; θ), θ ∈ Θ}, and develop the higher order asymptotic theory. The third-order Edgeworth expansion for a class of estimators is derived. It is shown that the maximum likelihood estimator is still third-order asymptotically optimal in our general situation. However, the Edgeworth expansion contains two terms which vanish in the case of curved exponential family. Regarding this point we elucidate some results which did not appear in Amari′s framework. Our results are applicable to time series analysis and multivariate analysis. We give a few examples (e.g., a family of curved ARMA models, a family of curved regression models).

Original languageEnglish
Pages (from-to)228-248
Number of pages21
JournalJournal of Multivariate Analysis
Issue number2
Publication statusPublished - 1994 Feb
Externally publishedYes


  • Curved ARMA models
  • Curved probability density
  • Curved regression model
  • Differential-geometrical method
  • Edgeworth expansion
  • Higher order asymptotic theory

ASJC Scopus subject areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty


Dive into the research topics of 'Statistical analysis of curved probability densities'. Together they form a unique fingerprint.

Cite this