APPLICATIONS OF ZERO INFLATED MODELS FOR HEALTH SCIENCES DATA

Abstract

Pneumonia is an infection of one or both lungs which is usually caused by bacteria, viruses, or fungi. Each year, pneumonia attack kills about 1.4 million people in the world, especially among children who are also the main sufferers of the disease. The aim of this study was to examine the factors that are associated directly or indirectly in pneumonia patients among the children. In this present paper, we have considered several regressions model to fit the count data that encounter in the field of Health Sciences. We have fitted Poisson, Negative Binomial (NB), Zero-Inflated Poisson (ZIP) and Zero-Inflated Negative Binomial (ZINB) regressions to pneumonia data. To compare the performance of these models, we analysed data with moderate to high percentage of zero counts. Because the variances were almost two times greater than the means, it appeared that both NB and ZINB models performed better than Poisson and ZIP models for the zero inflated and overdispersed count data. From the results of the ZINB regression can overcome overdispersion so it was better than the Poisson regression model.