IMPLEMENTASI METODE SINGULAR SPECTRUM ANALYSIS (SSA) PADA PERAMALAN INDEKS LQ45

Abstract

Stock market index forecasting is one of the research studies to obtain stock market index prediction using statistical and computational techniques, one of the techniques that can be used is Singular Spectrum Analysis (SSA). SSA aims to decompose a time series into a set of independent components which have nonstatic tendencies such as trends, oscillations, and noise. SSA can be applied without any prior assumptions such as stationarity, linearity, and normality which seem impossible to fulfill by the fluctuation of stock market index data. One of the stock indices listed in Indonesia Stock Exchange (IDX) is LQ45. The LQ45 index contains 45 selected stocks that have great performance in liquidity and fundamental aspects. LQ45 Index data has a nonstatic structure so it is suitable to be predicted using the SSA method. The forecasting uses the daily data of the LQ45 Index for the period of July 25, 2022 – November 14, 2022 with a window length (L) of 30. From the forecasting process, 4 groups were acquired and Mean Absolute Percentage Error (MAPE) value was 1.69% to forecast the next 20 days. These results indicate that the forecasting process using SSA for LQ45 Index of the period is accurate.