GARCH-BASED SECTOR ROTATION ON THE INDONESIA STOCK EXCHANGE: A SYSTEMATIC REVIEW

Abstract

The number of retail investors on the Indonesia Stock Exchange, mostly from younger generations, has grown rapidly, yet many of them still lack a structured, data driven investment strategy. This study aims to identify, evaluate, and synthesize the literature on multivariate generalized autoregressive conditional heteroskedasticity models, specifically Glosten Jagannathan Runkle and dynamic conditional correlation frameworks, for sector rotation strategies and portfolio backtesting on the Indonesia Stock Exchange Industrial Classification sectoral indices. A systematic literature review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta Analyses two thousand twenty protocol. Records were retrieved from Semantic Scholar through Publish or Perish, and two hundred sixty two records were screened down to one hundred forty six studies using inclusion and exclusion criteria derived from the Population Intervention Comparison Outcome Context framework. Five research questions guided the thematic synthesis. The results show that the leverage effect is consistently present in the Indonesian market and is well captured by the Glosten Jagannathan Runkle framework, that inter sectoral correlations are time varying and strengthen during economic crises, and that active model based strategies tend to outperform standard buy and hold approaches. Crucially, no reviewed study integrates Glosten Jagannathan Runkle models, dynamic conditional correlation frameworks, and a multifactor rotation score into a single walk forward backtesting framework for the eleven sectoral indices. This gap directly guides future research and the practical design of investment information systems that translate complex quantitative outputs into clear, actionable signals for individual investors.