Integrating Machine Learning into Portfolio Optimization: A Hybrid ARIMA–GARCH Predictive Model with Genetic Algorithm

Abstract

This article proposes a hybrid quantitative framework that integrates statistical time‑series modeling and evolutionary machine learning for portfolio optimization. The approach combines an ARIMA–GARCH model to jointly estimate conditional mean and volatility of asset returns with a Genetic Algorithm (GA) that searches for optimal portfolio weights on the basis of model‑implied return–risk profiles. The study adopts a secondary‑data quantitative design, synthesising evidence from prior empirical applications of ARIMA–GARCH forecasting and GA‑based portfolio optimization in equity markets, including the S&P500 index and Indonesian LQ45 constituents. Descriptive analysis confirms strong volatility clustering and leptokurtosis in daily stock index returns, justifying the use of GARCH‑type volatility models. Empirical results from the literature show that hybrid ARIMA–GARCH models significantly outperform standalone ARIMA and buy‑and‑hold strategies in terms of forecasting error and risk‑adjusted performance, while GA‑optimized portfolios achieve superior risk–return trade‑offs compared with traditional mean–variance optimization. These findings support the conceptual integration of ARIMA–GARCH forecasts and GA‑based allocation as a promising direction for portfolio construction, particularly in emerging markets such as Indonesia. The article concludes with implications for portfolio managers, regulators, and higher‑education curricula in quantitative finance and data‑driven investment.