Forecasting the Direction of Stock Price Movement Using Technical Indicator Inputs with Quantile Regression Neural Network (QRNN)
Rana Athaya Imtiyaz (a*), Dedy Dwi Prastyo (b), Irhamah (c)

Department of Statistics, Faculty of Science and Data Analytics, Institut Teknologi Sepuluh Nopember (ITS)
*ranaathayai[at]gmail.com

Abstract

The Indonesian stock market has experienced significant growth in investor participation from 2021 to 2024, with the banking sector emerging as a dominant force on the Indonesia Stock Exchange (IDX). Banking stocks such as BBCA (conventional banking), BRIS (Islamic banking), and ARTO (digital banking) exhibit diverse price movements, reflecting the unique dynamics of each banking type. Technical analysis plays a crucial role in forecasting stock price movements by leveraging historical price and trading volume data. This study applies the Quantile Regression Neural Network (QRNN) model, which integrates Quantile Regression (QR) to estimate price distributions at 5%, 50%, and 95% quantiles, and Neural Network (NN) to capture nonlinear relationships within the data. This approach enables the model to provide more precise predictions by modeling price fluctuations across different quantiles. Using time series data with forecasting horizons of 5, 10, and 22 trading days, the findings indicate that the QRNN model effectively forecasts stock price intervals. Performance evaluation based on Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) demonstrates the model^s optimal accuracy, making it a valuable tool for enhancing investment decision-making.

Keywords: Technical Indicators, Banking Stock Price Forecasting, QRNN, Time Series.

Topic: Financial management