Some numerical methods of rational characterization in causal time series models

  • Authors:

  • Concepción González-Concepción;María Candelaria Gil-Fariña;Celina Pestano-Gabino

  • Affiliations:

  • Department of Applied Economics, University of La Laguna, Tenerife, Canary Islands, Spain;Department of Applied Economics, University of La Laguna, Tenerife, Canary Islands, Spain;Department of Applied Economics, University of La Laguna, Tenerife, Canary Islands, Spain

  • Venue:
  • MATH'05 Proceedings of the 8th WSEAS International Conference on Applied Mathematics
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

The systematic study of data to obtain specific properties from long (or short) data series is a main objective. The use of rational models and related numerical methods can be useful to predict the behaviour of relevant economic variables. This paper is a continuation of González-Gil [16] which is concerned with illustrating the application of several numerical methods, among them, the corner method, epsilon-algorithm, rs-algorithm and qd-algorithm to time series modelling. These methods which are closely related to theoretical research in Padé Approximation have been proposed to identify some type of rational structure associated to economic data in different contexts (financial, marketing, farming...). Now, we present the study of the statistical significance for the four mentioned methods. Two examples will be considered, namely, a simulated ARMA model and a Transfer Function Model for the sales series M given in Box-Jenkins [7] and Tsay [27].