Comparative Analysis of Time Series Methods for Stock Market Index Forecasting

The capital market is a means by which companies or similar institutions can trade their various long-term financial instruments. One of the stocks that is much in demand by traders and investors is the stock owned by Bank Central Asia with the stock code of BBCA. BBCA shares during the period 2017 to 2019 always occupy position 2 in the ranking of stocks with the highest frequency of transactions in the banking sector. One of the problems faced by an investor in stock trading activities is volatility, where there will always be price fluctuations. This makes analysis activities very important as a determinant of success in stock trading. One form of analysis that can be done by stock investors is Technical Analysis, but the indicators used in technical analysis still cannot produce results that can describe price movements perfectly. The Time Series analysis method can be used to provide additional information to Technical Analysis because stock data is included in the time series data group. This study aims to find out the results of the implementation of three types of Time Series analysis methods, namely Linear Trend Regression, Exponential Smoothing, and ARIMA on Bank Central Asia stock price index data for the 2020-2022 period and determine the method with the best accuracy level among the three methods. The results of this study indicate that the ARIMA (1,1,0) method produces the smallest RMSE value at 162.9544 so that it has the best accuracy level among the three methods used. The findings of this study are expected to provide input to investors who want to invest capital into stock products and need additional analysis indicators in making investment decisions.