Machine Learning Prediction of Time Series Data (Decomposition and Forecasting Methods Using R)

Abstract

The machine learning prediction of time series data an analytical review explores the best way of time series machine learning analysis of two secondary sample data sets (air passenger and usgdp). Despite the fact that there were numerous types of analysis of time series tries to explore the best way, none of them would explore the suitable way to explore and predict the future train using r programming. Its principal object is to disclose the easiest method to analyze the time series whose information structure was a large association on other components. The intermediate outcomes were adequately e with explain with graphs and plots. The air passenger data was analyzed with acf ARIMA model for the forecast of future 10 years’ time found the increasing pattern. While figuring arima and pcf the bi-product BIC and AIC explain the pattern and their relationship with p-value were clarified with the fitting graphical conclusion of large data sets. Therefore, this document presents the easiest way to predict time series data and its significance for data prediction using r programming.