CEXPONENTIALLY WEIGHTED METHODS FOR FORECASTING INTRADAY TIME SERIES WITH MULTIPLE SEASONAL CYCLES

Abstract: Intraday data plays a vital role for Atmospheric Sciences for Wind Speed, Wind Wave length, Temperature etc. hourly data.In this paper, we have introduced two new Intraday data models i.e. New Exponential Smoothing and Trignometric model for intraday data. These two models are deduced according to seasons i.e. Summer, Winter and Rainy seasons. In this paper, two measures of accuracy are used. They are Mean Square error and Root Mean Square Error (RMSE). These two models are empirically tested using Atmospheric data of Gadanki India.

Keywords: New Exponential Smoothing, Intraday data, MSE, RMSE.