Base on Malaria count data report from 2000-2009 in Ubon Ratchathani province of north-eastern
Thailand, malaria incidence rates are computed by rain, mean temperature, minimum temperature,
maximum temperature, humidity and month. Linear regression model, Poisson and Negative binomial
GLM containing additive effects associated with the season of the year, climatic factors and the
malaria incidence rates in the previous months provides a good fit to the data, and can be used to
provide useful short-term forecasts. Although the season, rain, mean temperature, minimum
temperature, maximum temperature, humidity effects are all highly statistically significant, by far the
best predictor of the number of new cases occurring in any month is the disease incidence rate in the
preceding month. Having a model that provides such forecasts of disease outbreak.