Comparative Study of Concepts and Implementation of Various Univariate Time Series Analysis Methods

Adi Nugroho (Fakultas Teknologi Informasi Universitas Kristen Satya Wacana)

Indonesian Journal of Computing and Modeling
Vol 3 No 2 (2020)

Univariate time series analysis is a very important method in everyday life because this method has many practical applications. Various methods of time series analysis have been previously discovered by experts. However, until now the experts have not been able to determine exactly which method is best implemented in certain time series data. In this paper we will make a comparison of the 8 (eight) time series analysis methods that are most often used, namely SMA, EMA, WMA, SES, Holt's Method, Holt's Winter Seasonal Method, ARIMA, and SARIMA, with the aim of providing guidance to readers to choose which method is most appropriate for a given time series data. Research for this was conducted using the same data for all methods.