Finite-sample properties of the maximum likelihood estimator in autoregressive models with Markov switching

被引:28
作者
Psaradakis, Z [1 ]
Sola, M
机构
[1] Univ London Birkbeck Coll, Dept Econ, London W1P 2LL, England
[2] London Business Sch, Ctr Econ Forecasting, London NW1 4SA, England
[3] Univ Torcuato Tella, Dept Econ, RA-1428 Buenos Aires, DF, Argentina
关键词
finite-sample distribution; Markov-switching model; maximum likelihood estimator; regime shifts;
D O I
10.1016/S0304-4076(98)00010-4
中图分类号
F [经济];
学科分类号
02 ;
摘要
This paper examines the finite-sample properties of the maximum likelihood estimator in autoregressive models subject to Markov mean and variance shifts. Our results reveal that conventional asymptotic approximations to the distribution of the maximum likelihood estimator can often be poor for the sample sizes that are typical for annual and quarterly times series. (C) 1998 Elsevier Science S.A. All rights reserved.
引用
收藏
页码:369 / 386
页数:18
相关论文
共 44 条
[1]   BAYES INFERENCE VIA GIBBS SAMPLING OF AUTOREGRESSIVE TIME-SERIES SUBJECT TO MARKOV MEAN AND VARIANCE SHIFTS [J].
ALBERT, JH ;
CHIB, S .
JOURNAL OF BUSINESS & ECONOMIC STATISTICS, 1993, 11 (01) :1-15
[2]  
[Anonymous], J TIME SER ANAL, DOI DOI 10.1111/J.1467-9892.1986.TB00485.X
[3]  
CECCHETTI SG, 1990, AM ECON REV, V80, P398
[4]  
CHEN CWS, 1996, UNPUB UNIFIED APPROA
[5]  
Conover W. J., 1980, Practical nonparametric statistics, V2nd
[6]   SERIAL-CORRELATION IN LATENT DISCRETE VARIABLE MODELS [J].
COSSLETT, SR ;
LEE, LF .
JOURNAL OF ECONOMETRICS, 1985, 27 (01) :79-97
[7]  
Diebold F., 1994, Nonstationary Time Series Analysis and Cointegration, P283
[8]   Why are estimates of agricultural supply response so variable? [J].
Diebold, FX ;
Lamb, RL .
JOURNAL OF ECONOMETRICS, 1997, 76 (1-2) :357-373
[9]   DURATION-DEPENDENT TRANSITIONS IN A MARKOV MODEL OF UNITED-STATES GNP GROWTH [J].
DURLAND, JM ;
MCCURDY, TH .
JOURNAL OF BUSINESS & ECONOMIC STATISTICS, 1994, 12 (03) :279-288
[10]  
ENGEL C, 1990, AM ECON REV, V80, P689