Publication:
The exact likelihood function for the vector ARMA model

Loading...
Thumbnail Image
Official URL
Full text at PDC
Publication Date
1993
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Facultad de Ciencias Económicas y Empresariales. Instituto Complutense de Análisis Económico (ICAE)
Citations
Google Scholar
Research Projects
Organizational Units
Journal Issue
Abstract
This paper implements in Fortran 77 a new algorithm which has the same purpose as algorithm AS 242 of Shea (1989), namely to compute the exact likelihood function of a vector ARMA model. The new algorithm turns out to be faster in many relevant cases and not appreciably slower in any. In addition to advantages offered by the algorithm of Shea (1989), including the calculation of an appropiate set of residuals, it also permits the automatic detection of noninvertible models as a byproduct. The Fortran 77 code presented here combines improved versions of the algorithms due to Ljung and Box (1979) and Hall and Nicholls (1980) with an algorithm of Kohn and Ansley (1982). The resulting procedure puts together a set of useful features which can only be found separately in other existing methods.
En este trabajo se presenta la codificación en Fortran 77 de un nuevo algoritmo para evaluar la función de verosimilitud exacta de un modelo ARMA multivariante. Este algoritmo resulta significativamente más rápido que el SHEA (1982) en muchos casos, mientras que no es claramente más lento en ninguno. Además de proporcionar un vector de residuos apropiado, permite detectar, como subproducto, modelos no invertibles. El código es una combinación de los algoritmos de Ljung y Box (1979) y Hall y Nicholls (1980) mejorados, con un algoritmo de Kohn y Ansley (1982). Como resultado se obtiene un algoritmo con ciertas propiedades que sólo pueden encontrarse por separado en los procedimientos disponibles actualmente.
Description
Unesco subjects
Keywords
Citation
Dennis, J.E. and Schnabel, R.B. (1983) Numerical Methods for Unconstrainded Optimization and Nonlinear Equations. New Jersey: Prentice-Hall. Hall, A.D. and Nicholls, D.F. (1980) The Evaluation of Exact Maximum Likelihood Estimates for VARMA Models. J. Statist. Comput. Simul., 10, 251-262. Kohn, R. and Ansley, C.F. (1982) A Note on Obtaining the Theoretical Autocovariances of an ARMA Process, J. Statist. Comput. Simul., 15, 273-283. Ljung, G.M. and Box, G.E.P. (1979) The Likelihood Function of Stationary Autoregressive-Moving Average Models, Biometrika, 66, 265-270. Martin, R.S. and Wilkinson, J.H. (1965) Symmetric Decomposition of Positive Definite Band Matrices, Num. Math., 7, 355-361. Mauricio, J.A. (1993) Exact Maxinmm Likelihood Estimation of Stationary Vector ARMA models, I.C.A.E., D.T. Nº 9316. Moler, C.B. (1972) Algorithm 423: Linear Equation Solver, Commun. Ass. Comput. Mach., 15, 274. Nicholls, D.F. and Hall, A.D. (1979) The Exact Likelihood Function of Multivariate Autoregressive-Moving Average Models, Biometrika, 66, 259-264. Shea, B.L. (1984), Maximum Likelibood Estimation of Multivariate ARMA Processes via the Kalman Filter. In Time Series Analysis: Theory and Practice (ed. O.D. Anderson), vol. 5, pp. 91-101, Amsterdam: North-Holland. Shea, B.L. (1989) Algorithm AS 242: The Exact Likelihood of a Vector Autoregressive-Moving Average Model, Appl. Statist., 38, 161-204.