Portal de Periódicos da CAPES

Predicting binary time series of SO 2 using generalized additive models with unknown link function

2004; Wiley; Volume: 15; Issue: 7 Linguagem: Inglês

10.1002/env.687

1180-4009

Javier Roca‐Pardiñas, Wenceslao González‐Manteiga, Manuel Febrero–Bande, J. M. Prada‐Sánchez, Carmén Cadarso-Suárez,

Water Quality Monitoring and Analysis

Abstract The goal of article paper is to use the generalized additive model with unknown link function to predict the binary time series of SO 2 concentration. The authors propose a modified version of the local scoring algorithm that allows for the non‐parametric estimation of the link function, by using local linear kernel smoothers. Results derived from the simulation study and the application to real data reveal that the predictor obtained with the proposed model presents a better performance in comparison with those obtained by using alternative transformed binary regression models. Copyright © 2004 John Wiley & Sons, Ltd.