تجزیه و تحلیل رگرسیون پواسون برای بروز سیل

Currently, flood risk can be considered as the most serious threat, mainly in areas and countries where hardly any other natural risks occur. In order to analyze this kind of natural disaster, it is necessary to examine the long-term development of its occurrence around the world as well as its financial and other consequences. The main objective of the research was to apply Poisson regression on flood occurrence as the dependent variable. The set of explanatory variables under consideration was tested and subsequently the final model was determined. Poisson regression model, which is a generalized linear model, was chosen as a computing model. Using it guarantees consistent results when working with variables with non-normal data distribution (skewed and discrete). Thus OLS estimator cannot work and is replaced by MLE estimator. Consequently, confidence intervals of estimated parameters and all model results can be received. The research resulted in selecting the Poisson regression model with an estimated and significant parameter. Moreover, sample sets of selected countries were compared and evaluated in terms of overall intensity of flood occurrence.