Comparative Performance Analysis for Generalized Additive and Generalized Linear Modeling in Epidemiology

Bellali Hedia,Talmoudi Khouloud,Saez Marc,Ben-Alaya Nissaf,Malouche Dhafer,Chahed Mohamed

doi:10.14569/ijacsa.2017.081255

Bellali Hedia, Talmoudi Khouloud + Show 4 more

Open Access

https://doi.org/10.14569/ijacsa.2017.081255

Copy DOI

Abstract

Most environmental-epidemiological researches emphasize modeling as the causal link of different events (e.g., hospital admission, death, disease emergency). There has been a particular concern in the use of the Generalized Linear Models (GLMs) in the field of epidemiology. However, recent studies in this field highlighted the use of non-parametric techniques, especially the Generalized Additive Models (GAMs). The aim of this work is to compare performance of both methods in the field of epidemiology. Comparison is done in terms of sharpening the relation between the predictors and the response variable as well as in predicting outbreaks. The most suitable method is then adopted to elucidate the impact of bioclimatic factors on the emergence of the zoonotic cutaneous leishmaniasis (ZCL) disease in Central Tunisia. Monthly epidemiologic and bioclimatic data from July 2009 to June 2016 are used in this study. Akaike information criterion, R-squared and F-statistic are used to compare model performance, while the root mean square error is used for checking predictive accuracy for both models. Our results show the potential of GAM model to provide a better assessment of the nonlinear relations and to give a high predictive accuracy compared to GLMs. The results also stress the inaccurate estimation of risk factors when linear trends are used to model nonlinear structured data.

Highlights

In the last decade, there has been an increasing interest for the use of nonparametric modeling techniques in the field of epidemiology, especially the generalized additive models (GAMs) [1]
The most suitable method is adopted to elucidate the impact of bioclimatic factors on the emergence of the zoonotic cutaneous leishmaniasis (ZCL) disease in Central Tunisia
Our results show the potential of Generalized Additive Models (GAMs) model to provide a better assessment of the nonlinear relations and to give a high predictive accuracy compared to Generalized Linear Models (GLMs)

Summary

Introduction

There has been an increasing interest for the use of nonparametric modeling techniques in the field of epidemiology, especially the generalized additive models (GAMs) [1]. Researchers are still faithful to the use of parametric techniques such as the generalized linear models (GLM) [2] This can be explained by their robustness and the reliability of the results provided. Since their emergence, both approaches have been extensively applied in diverse domain such as environment, signal processing, ecology and in epidemiology [3]-[5]. This can be explained by their ability to describe the real dynamics existing in the data and to the straightforward way of interpreting and representing the results using graphical ways

Objectives

Methods

Results