NEGATIVE BINOMIAL REGRESSION

Dengue disease is one of the vector borne diseases in the world which records higher number of mortality and morbidity rates. In Sri Lanka, it is one of the leading diseases where public domain has a major concern. Innovative strategic plans should be implemented to reduce the impact of the disease. Availability of an effective and representative model will be useful in taking actions towards controlling the disease. With this broad aim, the objective of this study is to model the reported dengue cases in Colombo, Sri Lanka which is the commercial capital of Sri Lanka. Data consists of weekly reported dengue cases, weekly average maximum temperature, weekly average minimum temperature and weekly average rainfall in Colombo district from 43rd week of 2009 to 52nd week of 2016. Data from 2009 to 2015 were served for the development of the model and rest of the data for the validation purpose. Cross-correlation analysis was applied to track the movements of mentioned climatic variables with the dengue series. Negative binomial regression technique was served to achieve the objective of the study.