Asymptotic distribution of maximum likelihood estimators in linear models with autoregressive disturbances

Clifford Hildreth

ResearchPublished 1966

Some statistical properties are established for maximum likelihood estimates of linear model coefficients whose disturbances are generated by a stationary linear first-order autoregressive process with unknown autoregression coefficient. This study describes the model, examines the likelihood function, and gives a detailed proof that the estimates of coefficients of independent variables and the estimate of the autoregression coefficient have a limiting joint multivariate-normal distribution, with the estimate of autoregression distributed independently of the estimates of coefficients of independent variables. This asymptotic covariance matrix of the latter estimates is the same as that of the best linear unbiased estimates for a model in which the autoregression coefficient is known. Up to now, consistency is the only property that has been shown for maximum likelihood estimates. Results of this study should be especially helpful in large-sample applications. 30 pp.

Document Details

  • Copyright: RAND Corporation
  • Availability: Web Only
  • Year: 1966
  • Pages: 30
  • Document Number: RM-5059-PR

Citation

Chicago Manual of Style

Hildreth, Clifford, Asymptotic distribution of maximum likelihood estimators in linear models with autoregressive disturbances. Santa Monica, CA: RAND Corporation, 1966. https://www.rand.org/pubs/research_memoranda/RM5059.html.
BibTeX RIS

This publication is part of the RAND research memorandum series. The research memorandum series, a product of RAND from 1948 to 1973, included working papers meant to report current results of RAND research to appropriate audiences.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.