Big Data Analytics for Inflation Forecasting: Integrating Alternative Data and Islamic Economic Models

Abstract

This study aims to develop an integrated inflation forecasting framework by combining big data analytics with Islamic economic models to enhance predictive accuracy and policy relevance. Conventional inflation forecasting methods often rely on limited macroeconomic indicators and overlook the ethical and behavioral dimensions emphasized in Islamic economics. To address this gap, the research incorporates alternative data sources, including online price indices, digital transaction records, and real-time consumer sentiment indicators, alongside Islamic economic variables such as profit-and-loss-sharing dynamics, zakat-related liquidity flows, and principles of consumption moderation. The study adopts a quantitative big data analytics approach, employing machine learning techniques to process high-volume, high-velocity, and high-variety datasets. Multiple forecasting models were developed and compared, including baseline econometric models, standalone big data-driven models, and hybrid models integrating Islamic economic constructs. Model performance was evaluated using standard forecasting accuracy metrics across multiple time horizons. The results demonstrate that the hybrid models significantly outperform conventional approaches in terms of predictive accuracy and robustness, particularly during periods of economic uncertainty and price volatility. The inclusion of alternative data improves the timeliness of inflation signals, while Islamic economic variables enhance the stability and interpretability of the forecasts. These findings suggest that integrating big data analytics with Islamic economic perspectives offers a more comprehensive and resilient framework for inflation forecasting. The study provides practical implications for policymakers, central banks, and Islamic financial institutions seeking data-driven yet value-based approaches to macroeconomic stabilization.