Research Information System
in: Artificial Intelligence and the Future of Islamic Finance, Taylor & Francis Ltd, pp.262-272, 2026
Inflation inertia manifests as a phenomenon in economies wherein a heightened or diminished inflation rate tends to endure, exhibiting a reluctance to promptly adjust to shifts in economic conditions. This phenomenon poses a significant challenge for central banks and economic policymakers. As inflation inertia intensifies, the task of steering the economy and maintaining price stability becomes increasingly formidable. Following the Russia-Ukraine conflict, a surge in inflation occurred across both developed and developing nations, driven by escalating energy commodity prices. The resurgence of the inflation inertia problem has sparked renewed debate and concern among experts and policymakers. In this research, we examine the determinants of inflation inertia using a combination of traditional time series models and advanced supervised machine learning techniques. Initially, we gauge inflation inertia through the application of dynamic Kalman Filter models. Subsequently, we scrutinize the factors influencing inflation inertia by incorporating macroeconomic and financial indicators such as unemployment, commodity prices, output gap, 10-year government bond yield, and more, spanning the period from January 2003 to October 2023. Within this framework, we employ both conventional time series models and sophisticated supervised machine learning models. A comprehensive comparison of the forecast performance of these alternative models aims to determine the superior model for accurately estimating inflation inertia. The outcomes of this analysis hold the potential to furnish valuable insights for policymakers, offering guidance on the integration of artificial intelligence-based models in macroeconomic forecasts.