Discussiones Mathematicae Probability and Statistics 23(2) (2003) 147-165

AUTOREGRESSIVE ERROR-PROCESSES, CUBIC SPLINES AND TRIDIAGONAL MATRICES

Hilmar Drygas

Universität Kassel
Fachbereich 17 Mathematik/Informatik
Heinrich-Plett-Straße 40, D-34132 Kassel
e-mail: drygas@mathematik.uni-kassel.de

Abstract

In the paper formulate for the inversion of some tridiagonal matrices are given. The results can be applied to the autoregressive processes.

Keywords: autoregressive processes, cubic splines interpolation, linear regression model, time series.

2000 Mathematics Subject Classification: 62M10.

References

[1]H. Drygas, Sufficiency and completeness in the general Gauss-Markov model. Sankhya Ser A 45 (1984), 88-98.
[2]H. Drygas, The inverse of a tridiagonal matrix. Kasseler Mathematische Schriften, forthcoming (2003).
[3]F. Graybill, Matrices with Applications in Statistics. Wadsworth & Brooks/Cole, Advanced Books & Software, Pacific Grove, California 1963.
[4]R. Nabben, Two sided formels on the inverses of diagonally dominant tridiagonal matrices. Linear Algebra and its Applications 287 (1999), 289-305.
[5]C.R. Rao, Linear statistical inference and its applications. John Wiley & Sons Inc., New York-London-Sydney-Toronto 1973.
[6]P. Schönfeld, Methoden der Ökonometrie. Band I, Verlag Franz Vahlen GmbH, Berlin und Frankfurt am Main 1969.
[7]H.R. Schwarz, Numerische Mathematik. R.G. Teubner-Verlag, Stuttgart 1988.
[8]J.Stoer, Numerische Mathematik 1. 5. Auflage, Springer-Verlag, Berlin Heidelberg New York London Paris Tokyo Hong Kong 1989.
[9]W. Törnig, P. Spellucci, Numerische Mathematik für Ingenieure und Physiker. Band 2: Numerische Methoden der Analysis. Springer Verlag, Berlin Heidelberg New York London Paris Tokyo Hong Kong 1990.

Received 10 February 2003