An Approach to Study on MA, ES, AR for Sunspot Number (SN) Prediction and to Forecast SN with Seasonal Variations Along with Trend Component of Time Series Analysis Using Moving Average (MA) and Exponential Smoothing (ES)

Abstract

Sunspots are the interesting things on the surface of sun which is why it would be more engaging if sunspots become predictable. In this study, sunspot numbers (SN) have been predicted within recent solar cycle 24. To find the best model, moving average (MA), exponential smoothing (ES) and auto regression (AR) have been used. Besides these, in another two experiments which are seasonal component was extracted from data using moving average (MA) and exponential smoothing (ES) as well as with the help of simple regression analysis (RA), trend component was calculated. This exploration is entirely about understanding the differences among these models and the influences of those two components to predict sunspot using moving average (MA) and exponential smoothing (ES). Prediction results have been conducted to expose that difference and influence. It can manifest the way of forecasting for other model of time series analysis (TSA) to predict sunspot number (SN).