The Fourier series estimator to predict the number of dengue and malaria sufferers in Indonesia

In statistical modeling, Fourier series estimator is frequently applied to time series data in nonparametric approach. Based on nonparametric regression study, Fourier series estimator has good flexibility to predict seasonal and the combination of trend and seasonal data pattern. This paper proposes an application of Fourier series estimator in biostatistics and epidemiology cases. One of the important problems in health science is disease prevention efforts. For prevention efforts, prediction the number of sufferers is determined. Fourier series estimator is applied to predict the number of sufferers for seasonal diseases like dengue fever and malaria that becomes a main issue in Indonesia. We used secondary data from the Ministry of Health of Indonesia to model and predict the number of dengue fever and malaria sufferer in Indonesia based on Fourier series estimator. A selected model that be chosen has met the goodness of model's criteria such as the small Generalized Cross Validation (GCV), Mean Square Error (MSE), and the large determination coefficient. The selected model also considered the model parsimony. Therefore, Fourier series estimator was able to predict the number of sufferers for seasonal diseases and the prediction produced small value of Mean Absolute Percentage Error (MAPE) and MSE. Thus, the result can be used to give recommendation to related policy maker.