<![CDATA[The Composite Stock Price Index (IHSG) is a of the key indicator a country uses to assess its economic condition. The fluctuating movements of stock prices create uncertainly in the stock market, complicating decision-making for investors and government entities. Therefore, there is a need for a method that can forecast the Composite Stock Price Index to monitor such fluctuations. The objective of this study is to model the Composite Stock Price Index Utilizing a hybrid method and to assess the accuracy of this hybrid approach. The hybrid method employed is the Autoregressive Fractionally Integrated Moving Average (ARFIMA)-Artificial Neural Network (ANN). The results of this study show that the best ARFIMA model is ARFIMA (1,d,1) with a differencing parameter of dR/S = 0,362. The ANN model’s optimal architecture obtained through the backpropagation algorithm is ANN (3,2,1). The accuracy of the hybrid ARFIMA-ANN model, measured by the Mean Absolute Percentange Error (MAPE), yielded of 1,0164%, lower than the MAPE value of 1,7326% for the standalone ARFIMA model. This suggests that the hybrid model improves forecasting accuracy and is the most efferctive model for predicting the IHSG. ]]>
Copyrights © 2024
