1- Ferdowsi University
2- Ferdowsi University ,otfalipour@um.ac.ir
2- Ferdowsi University ,
Abstract: (75 Views)
Inflation, as a key indicator of economic performance, directly affects household purchasing power, price stability, and the long-term planning of firms and governments. Uncontrolled inflation not only reduces public welfare and exacerbates social inequalities, but also disrupts investment and sustainable growth by creating instability in economic expectations. Therefore, identifying the drivers of inflation and understanding their transmission mechanisms is essential for designing effective monetary and fiscal policies. This study investigates the impact of economic shocks on the inflation rate by employing a combination of two approaches: first, the random forest-based variable selection method with recursive feature elimination (RF-RFE) to identify the most influential factors, and second, the Bayesian vector autoregression (BVAR) model to analyze the time dynamics and mutual interactions of these shocks.The dataset covers 42 economic variables from the first quarter of 2009 to the fourth quarter of 2021, grouped into seven categories: supply, demand, monetary and banking, taxation and budget, exchange rate, energy, and employment. In the first step, the RF-RFE method identified the most important determinants of consumer inflation. The results indicated that five key variables producer inflation, value added in the oil and gas sector, quasi-money, the market exchange rate, and banknotes and coins in circulation play a major role in explaining consumer price fluctuations.The subsequent BVAR analysis showed that shocks originating from producer inflation and the exchange rate exert strong short-term effects on consumer inflation. By contrast, variables such as oil and gas value added play a moderating role in the long term, gradually alleviating inflationary pressures. Furthermore, the variance decomposition of forecast errors suggests that, in the long run, exchange rate volatility and liquidity changes driven by quasi-money increasingly account for fluctuations in inflation
Keywords: Inflation, Random Forest with Recursive Feature Elimination (RF-RFE), Bayesian Vector Autoregression (BVAR)
Type of Study: توسعه ای |
Subject:
رشد و توسعه و سیاست های کلان
Received: 2025/09/27 | Accepted: 2026/04/22
Received: 2025/09/27 | Accepted: 2026/04/22
Send email to the article author
| Rights and permissions | |
 | This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
