Time Series Forecasting Using Statistical And Neural Networks Models

Forecasting is a common statistical task in many areas where it contributes to inform decisions about the scheduling of production transportation personnel etc. And it provides a guide to long-term strategic planning. In many areas such as financial energy economics the time series data are non-stationary contain trend and seasonal variations. The goal of this thesis is to forecast the time series using two approaches namely the statistical approaches; they are seasonal ARIMA seasonal VARIMA models and Neural Networks approach and compare them in order to find the best model for time series forecasting. The energy area has an important role in the development of countries; thus consumption planning of energy must be made accurately despite they are governed by other factors such as that population gross domestic product weather vagaries storage capacity etc.