Department of Statistics, University of Ibadan, Ibadan, Nigeria; Department of Mathematical Science, Olabisi Onabanjo University, Ago-Iwoye, Nigeria
Ojo, J.F., Department of Statistics, University of Ibadan, Ibadan, Nigeria; Olatayo, T.O., Department of Mathematical Science, Olabisi Onabanjo University, Ago-Iwoye, Nigeria
Significant effort have been made in the study of the theory of subset autoregressive models and subset autoregressive moving average models; but less concerted effort have been made in the theory of subset autoregressive integrated moving average models. This study therefore focuses on the estimation and performance of subset autoregressive integrated moving average models. In this study, subset autoregressive integrated moving average models were compared with full autoregressive integrated moving average models. The parameters of these models were estimated using Marquardt algorithm and Newton-Raphson iterative method and the statistical properties of the derived estimates were investigated. An algorithm was proposed to eliminate redundant parameters from the full order autoregressive integrated moving average models. To determine the performance of the models, residual variance, Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were adopted. Results revealed that the residual variance attached to the subset autoregressive integrated moving average models is smaller than the residual variance attached to the full autoregressive integrated moving average models. Subset autoregressive integrated moving average models performed better than the full autoregressive integrated moving average models. © EuroJournals Publishing, Inc. 2009.
