New Exponential Smoothing For Intraday Data

Source abstract

Now-a-days intraday time series models are playing a wide role in various sectors like wind speed data, temperature data, sensex data etc. In this paper, the authors have introduced a new exponential smoothing model and Auto Regressive Moving Average (ARMA) models for intraday data. These two models are empirically tested using wind speed data of Gadanki, Chittoor District. Between these two models which model is better performed and the wind speed data is tested using Root Mean Square Error (RMSE). Kolmogorov Smirnov test is used for goodness of fit. Among the compared 25 ARMA models and new exponential smoothing model, RMSE of new exponential smoothing model is low. Therefore, the authors have concluded that, the new exponential smoothing model is the best model for wind speed data of Gadanki.