Multiple Optima and Asymptotic Approximations in the Partial Adjustment Model

Posted: 26 Oct 2008

See all articles by Douglas A. McManus

Douglas A. McManus

Federal Home Loan Mortgage Corporation (FHLMC)

John C. Nankervis

University of Essex - Department of Accounting, Finance & Management

N. Eugene Savin

University of Iowa - Henry B. Tippie College of Business - Department of Economics

Date Written: October 13, 2008

Abstract

This paper examines statistical problems which arise in empirical applications of the partial adjustment model with autoregressive errors when the model is nearly nonidentified. The results of Monte Carlo experiments show that the NLS estimation criterion function is multipeaked with high probability when the model is nearly nonidentified. In the cases examined the finite-sample distributions of the NLS estimators and the Wald test statistics are poorly approximated by their asymptotic distributions. The asymptotic approximation works better for the likelihood ratio (LR) test statistics, but still can be unsatisfactory. When the Wald and LR tests are based on bootstrap critical values the size distortions are effectively eliminated.

Keywords: Econometrics, Asymptotic Theory, Power

JEL Classification: C12, C13

Suggested Citation

McManus, Douglas A. and Nankervis, John C. and Savin, Nathan Eugene, Multiple Optima and Asymptotic Approximations in the Partial Adjustment Model (October 13, 2008). Journal of Econometrics, Vol. 62, 1994, Available at SSRN: https://ssrn.com/abstract=1284108

Douglas A. McManus (Contact Author)

Federal Home Loan Mortgage Corporation (FHLMC) ( email )

8200 Jones Branch Road
McLean, VA 22101
United States
703-903-2953 (Phone)

John C. Nankervis

University of Essex - Department of Accounting, Finance & Management ( email )

Wivenhoe Park
Colchester CO4 3SQ
United Kingdom

Nathan Eugene Savin

University of Iowa - Henry B. Tippie College of Business - Department of Economics ( email )

108 Pappajohn Building
Iowa City, IA 52242
United States
319-335-0855 (Phone)

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
473
PlumX Metrics