Abstract

PARAMETER ESTIMATION OF LOG-ARIMAX MODEL VIA GENERALIZED LINEAR METHOD OF EXPONENTIAL FORM

K. I. EKERIKEVWE* and T.O. OLATAYO

---

To improve forecast accuracy, an autoregressive integrated moving average model, ARIMAX (p, d, q, b), was developed for short-memory observational time series data with exogenous covariate(s) .However, for long-memory frequency observations, a modification will be necessary to neutralize the model for a better and improved prediction of the system. This study, therefore, is designed to propose and formulate a logarithmic autoregressive integrated moving average (LOG-ARIMAX)modelwhose distributional form would be robust and sufficient in capturing and accommodating both the external covariate (s) and the heavy-tailed properties of long-memory frequencyobservational time series data. The parameter estimation of the LOGARIMAX model will be carried out via Generalized Linear Method (GLM)of exponential form.The comparison of the model performance indexes will bedone with the traditional ARIMAX model under in-sample forecasts conditions. Keyword: ARIMAX, Time Series, Accuracy, Frequency, Model, Estimation

PDF