Global Ease of Living Index: a machine learning framework for longitudinal analysis of major economies

Researchers have published a machine learning framework that constructs a Global Ease of Living Index, tracking quality of life across major economies from 1970 onward by combining socio-economic and infrastructural indicators into a single composite score [1]. The framework, detailed in a paper by Tanay Panat and Rohitash Chandra, uses dimensionality reduction techniques—Principal Component Analysis and Factor Analysis—to distill a wide range of economic indicators into a manageable set of factors [1][4]. The authors address the problem of missing data for certain indicators in specific countries through machine learning imputation before curating and updating the dataset [1][2]. The resulting index is a weighted sum of four sub-indices: Economic, Institutional, Quality of Life, and Sustainability [4]. The Quality of Life sub-index receives the highest weight at 0.35, while the Economic and Institutional sub-indices are each weighted at 0.25, and Sustainability at 0.15 [4]. The weighting scheme draws partial inspiration from the Ease of Living Index created by the Government of India, though the new global index places less reliance on economic factors and greater emphasis on institutional and quality-of-life measures such as healthcare and government effectiveness [4]. The study ranks countries annually from 1970 through 2021, focusing on G20 members and comparing nations at different development stages [4][5]. The authors state the tool is designed to help policymakers identify areas needing targeted intervention, including healthcare systems, employment opportunities, and public safety [1][2]. The code and data are released openly to enable reproduction and application in other contexts [1][2].