Statistical Modelling of Moderate Distribution in Resource Allocation
Abstract
Moderate disparities in resource allocation were often overlooked in favour of extreme inequality, despite their long-term implications for fairness, access, and welfare. In the present study, the factors associated with moderate distribution patterns were examined, statistical models were constructed to predict moderate distribution outcomes, and alternative allocation strategies were evaluated. A simulated cross-sectional dataset of 60 administrative regions was created for illustrative empirical analysis. Resource allocation balance was operationalised through a Moderate Distribution Score (MDS), while explanatory variables included social need, poverty intensity, environmental vulnerability, institutional capacity, and population pressure. Correlation analysis, multiple linear regression, and random forest regression were applied. Allocation strategies were then compared through simulated policy scenarios involving equal-share, population-based, need-based, and hybrid weighted allocation rules. It was found that institutional capacity, social need recognition, and poverty intensity were positively associated with moderate distribution outcomes, whereas population pressure and environmental vulnerability exerted adverse effects. In predictive terms, random forest regression was found to outperform linear regression, although both models demonstrated acceptable forecasting ability. Among the strategy scenarios, the hybrid weighted model was found to produce the strongest balance between equity, sustainability, and efficiency, followed by the need-based model. It was concluded that moderate distribution outcomes were shaped by a combination of socioeconomic and institutional factors and that multi-criteria allocation strategies were better suited than simple equal-share rules for promoting equitable resource allocation.
How to Cite
Vijay Laxmi, Rachna Khandelwal, "Statistical Modelling of Moderate Distribution in Resource Allocation", Vol. 3, Issue 9, 18-12-2025, pp. 57-68.
