Predicting Malaria Cases in Sudan Based on Time Series analysis

Abstract

Malaria is one of the most prevalent and debilitating diseases afflicting humans in Africa, and it is responsible for the fifth greatest number of deaths due to infectious diseases in the world and it is from the most ten infectious diseases that causing deaths In Sudan. That’s why the study of infectious diseases represents one of the richest areas in mathematical biology. The infectious diseases need fast responses, and appropriately modeling, also predicting the outcome of disease spread over time and across space is a critical step toward informed development of effective strategies for public health intervention and decision making. Using mathematical representations for infectious diseases let the essential elements grasped quickly, and captured well. A lot of difficulties face the responsible organization in the process of collecting infectious diseases data by traditional ways in addition to the presence of more than 1500 health center in Sudan, as well; there is no applied method in Sudan to predict the spread of infectious diseases. Determine the best and most efficient mathematical model for predicting the new cases of infectious diseases in Khartoum, Al-Gadaref and Sennar based on the previous (history) data and visualize diseases distribution development (ArcMap) is important thing to save lives. Using time series analysis (Auto Regressive model, Moving Average model, Mixed Model, and Exponential Smoothing model) found that the simple models (AR and MA) represented the data better in Khartoum, Al-Gadaref and Sennar states in the seasonal data and Sennar in the non-seasonal data. While the Exponential Smoothing and Mixed model (ARIMA) are better in representing Khartoum and Al-Gadaref non-seasonal data. Which prove that; not all data can be representing using the same forecasting model.