Likelihood and parametric heteroscedasticity in normal connected linear models

Joao Tiago Mexia; Pedro Corte Real

Discussiones Mathematicae Probability and Statistics (2000)

  • Volume: 20, Issue: 2, page 177-188
  • ISSN: 1509-9423

Abstract

top
A linear model in which the mean vector and covariance matrix depend on the same parameters is connected. Limit results for these models are presented. The characteristic function of the gradient of the score is obtained for normal connected models, thus, enabling the study of maximum likelihood estimators. A special case with diagonal covariance matrix is studied.

How to cite

top

Joao Tiago Mexia, and Pedro Corte Real. "Likelihood and parametric heteroscedasticity in normal connected linear models." Discussiones Mathematicae Probability and Statistics 20.2 (2000): 177-188. <http://eudml.org/doc/287719>.

@article{JoaoTiagoMexia2000,
abstract = {A linear model in which the mean vector and covariance matrix depend on the same parameters is connected. Limit results for these models are presented. The characteristic function of the gradient of the score is obtained for normal connected models, thus, enabling the study of maximum likelihood estimators. A special case with diagonal covariance matrix is studied.},
author = {Joao Tiago Mexia, Pedro Corte Real},
journal = {Discussiones Mathematicae Probability and Statistics},
keywords = {linear model; connected model; normal model; maximum likelihood estimators; score function; Newton-Raphson method; connected models},
language = {eng},
number = {2},
pages = {177-188},
title = {Likelihood and parametric heteroscedasticity in normal connected linear models},
url = {http://eudml.org/doc/287719},
volume = {20},
year = {2000},
}

TY - JOUR
AU - Joao Tiago Mexia
AU - Pedro Corte Real
TI - Likelihood and parametric heteroscedasticity in normal connected linear models
JO - Discussiones Mathematicae Probability and Statistics
PY - 2000
VL - 20
IS - 2
SP - 177
EP - 188
AB - A linear model in which the mean vector and covariance matrix depend on the same parameters is connected. Limit results for these models are presented. The characteristic function of the gradient of the score is obtained for normal connected models, thus, enabling the study of maximum likelihood estimators. A special case with diagonal covariance matrix is studied.
LA - eng
KW - linear model; connected model; normal model; maximum likelihood estimators; score function; Newton-Raphson method; connected models
UR - http://eudml.org/doc/287719
ER -

References

top
  1. [1] T. Amemiya, Advanced Econometrics, Harvard University Press, Harvard 1985. 
  2. [2] E.R. Berndt B.H. Hall and J.A. Hausman, Estimation and Inferencein Nonlinear Structural Models, Annals of Economic and Social Measurement 3 (1974), 653-666. 
  3. [3] W.C. Davidson, Variable Metric Methods for Minimization, Atomic Energy Commision, Research Development Report ANL-5990, Washington, D.C. 1959. 
  4. [4] S.M. Goldfeld, R.E. Quandt and H.F. Trotter, Maximization by Quadratic Hill-Climbing, Econometrica 34 (1966), 541-551. 
  5. [5] S.M. Goldfeld and R.E. Quandt, Nonlinear Methods in Econometrics,North-Holland Publishing, Amsterdam 1972. Zbl0231.62114
  6. [6] R.I. Jennrich, Asymptotic Properties of Non-Linear Least-Squares Estimators, The Annals of Mathematical Statistics 40 (1969), 633-643. Zbl0193.47201
  7. [7] J.D. Jobson and W.A. Fuller, Least Squares Estimation When the Covariance Matrix and Parameter Vector Are Functionally Related, Journal of the American Statistical Association 75 (1980), 176-181. Zbl0437.62064
  8. [8] J.L. Kelley, General Topology, Princeton, New Jersey 1961. Zbl0157.53002
  9. [9] J.T. Mexia, Linear Models with Partially Controlled Heteroscedasticity, Trabalhos de Investigaçao 2. Departamento de Matemática - Faculdade de Ciencias e Tecnologia/Universidade Nova de Lisboa 1993. Zbl0806.62054
  10. [10] R.L. Plackett, Principles of Regression Analysis, Oxford University Press, Oxford 1960. Zbl0091.31301
  11. [11] S.J. Prais and H.S. Houthakker, The Analysis of Family Budgets, Cambridge University Press, Cambridge 1955. 
  12. [12] C.R. Rao, Linear Statistical Inference and Its Applications (Second Edition), John Wiley & Sons 1973. Zbl0256.62002

NotesEmbed ?

top

You must be logged in to post comments.