Forecasting performance of logistic STAR model: An alternative version to the original LSTAR models

Article type: Research Article

Authors: Adebile, O.A.^a | Shangodoyin, O.A.^{b; *} | Arnab, R.^c

Affiliations: [a] University of Ibadan, Nigeria | [b] University of Swaziland, Swaziland | [c] University of Botswana, Botswana

Correspondence: [*] Corresponding author: Dr. D.K. Shangodoyin, Department of Statistics, University of Swaziland, P/Bag 4, Kwaluseni, Swaziland. E-mail: shangodoyind@ukzn.ac.za

Abstract: This paper proposes an alternative representation of the original version of the Logistic (LSTAR) model. The LSTAR and ESTAR models are frequently used in empirical research. The LSTAR model describes asymmetrical nonlinear adjustment process, while the ESTAR model describes symmetrical nonlinear adjustment process [19] in Liews [10]. To Liews et al., the theoretical assumption that exchange rate adjustment is symmetric makes the LSTAR model inappropriate for modelling exchange rate movements. They claim that this is the reason the LSTAR model has being neglected in the modelling of exchange rate in the past. They proposed the ALSTAR model. This version they noted allows a V-shape symmetric adjustment in exchange rate behaviour. In this paper, we also propose a similar but better alternative to their model. The square logistic smooth transition auto-regressive model (SLSTAR) has a shape exactly similar to the inverted bell-shape of the exponential model which allows a symmetric adjustment in exchange rate behaviour. Simulation studies were carried out to investigate the sample performance of our model. We also demonstrated our model with life data.

Keywords: Non-linearity, LSTAR ALSTAR, SLSTAR models

DOI: 10.3233/MAS-2006-1302

Journal: Model Assisted Statistics and Applications, vol. 1, no. 3, pp. 139-146, 2006