The Effect of Misery Index on Income inequality in Iran: Application of Machine Algorithms and Deep Learning

Kian Poor, Saeed; Darbidi, Maryam; Shamsollahi, Reza; Bayat, Faezeh

doi:10.30465/ce.2025.50265.1991

The Effect of Misery Index on Income inequality in Iran: Application of Machine Algorithms and Deep Learning

Document Type : Scientific-research

Authors

saeed kian poor ¹

maryam Darbidi ²

reza shamsollahi ²

Faezeh Bayat ³

¹ member faculty of economic/payam noor

² Economic

³ Master student of Economics, University of Isfahan

10.30465/ce.2025.50265.1991

Abstract

Income inequality remains a critical issue in Iran’s economy, exacerbating poverty, reducing investment, and threatening social stability, necessitating a thorough investigation of its determinants. This study examines the impact of the misery index (a composite of inflation and unemployment) on income inequality in Iran over the period 1981–2022, employing machine learning and deep learning algorithms implemented via Python and Google Colab, with the Shapley Additive Explanations (SHAP) method for data analysis. Findings reveal that the misery index, foreign direct investment, economic openness, and government size positively and significantly affect income inequality, increasing the income gap, while urban population growth has a negative effect, contributing to reduced inequality. Given Iran’s unique economic conditions, including high inflation and sanctions, the study recommends that the government implement supportive policies such as raising the minimum wage, providing targeted subsidies to lower-income deciles, and creating sustainable employment opportunities to reduce income inequality. These measures can enhance the purchasing power of low-income groups and promote social justice.

Keywords

: Misery Index

Income Inequality

Machine Learning Algorithms

Shapley Method

Economic Policy

Subjects