PREDICTING BBRI STOCK PRICES USING ARIMA MODEL AND CROSS-CORRELATION TECHNIQUE WITH STATE-OWNED BANK STOCKS

• Main Author: Kuswanto, Amanda
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/81426
• Institution: Institut Teknologi Bandung

Time series analysis is a statistical method used to understand the characteristics of data over time. In financial time series analysis, there are several approaches such as ARIMA models and cross-correlation analysis that can aid in predicting stock prices. The final project endeavors to model the stock prices of PT Bank Rakyat Indonesia (BBRI) using time series analysis and cross-correlation techniques. The initial method involves predicting BBRI stock prices with an ARIMA model, and the best model identified is ARIMA(4,1,4). The selection of the best model is based on the principle of model parsimony, diagnostic test results, and the lowest MAPE value. The MAPE value serves as a criterion for model selection as it can indicate the model’s performance in prediction. In addition to time series analysis, this final project also utilizes cross-correlation analysis to explore the relationship between BBRI stock prices and other bank stocks in the same index, namely BMRI, BBNI, and BBTN. Cross-correlation analysis shows that the linear relationship and causal relationship between BBRI stock prices and other bank stock prices can be examined in the form of a multiple linear regression equation. Several assumptions are used in multiple linear regression, including linearity, residual normality, residual independence, residual homoscedasticity, and model significance. These assumptions will be tested by residual plots, QQ normal plots, Durbin-Watson test, Breusch-Pagan test, and F test. By using the cross-correlation analysis, it is hoped to gain a deeper understanding of the correlation between these stocks, thus providing insights for investors in investment decision-making. This final project focuses on the importance of modeling and statistical analysis in understanding stock market dynamics and generating more effective investment strategies.