Interval Forecasting of Stock Price Movements with Technical Indicator Input using Hybrid Quantile Regression and SVR Angger Salsabila Rufida*, Dedy Dwi Prastyo, Tintrim Dwi Ary Widhianingsih
Department of Statistics, Faculty of Science and Data Analytics, Institut Teknologi Sepuluh Nopember (ITS)
*anggersalsabila21[at]gmail.com
Abstract
Stock investment is recognized as a strategic means to achieve long-term economic stability. As instruments representing ownership in companies, stocks offer high liquidity and substantial profit potential. Within the Indonesia Stock Exchange (IDX), the banking sector particularly Bank Central Asia (BBCA), Bank Negara Indonesia (BBNI), Bank Rakyat Indonesia (BBRI), and Bank Mandiri (BMRI) has consistently maintained positions within the IDX30, an index comprising highly liquid stocks with large market capitalizations. Accurate stock price forecasting plays a critical role in supporting informed investment decisions. Technical analysis, which leverages historical price data and trading volume, remains a widely adopted approach for anticipating market trends. This study proposes a hybrid predictive model utilizing various technical indicators as input features, including Moving Average (MA), Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), Price Rate of Change (ROC), Stochastic Oscillator, Williams %R, and Commodity Channel Index (CCI). The experimental results demonstrate that the hybrid model enhances prediction accuracy. Support Vector Regression (SVR) effectively captures the non-linear dynamics of stock price movements, while Quantile Regression (QR) enables the estimation of predictive intervals. The combination of these methods not only improves forecasting precision but also supports investors in developing risk-aware strategies and making well-informed investment decisions.
Keywords: Stock, Technical Indicators, Quantile Regression, Support Vector Regression
