Abstract
To diagnose the factors affecting poverty in (11) selected developing countries during the period (2000-2020), a regression was made for the poverty rate (pov), as a dependent variable on (8) explanatory variables, including the time variable (t) affecting poverty for the study sample. Using three static regression models using (Panel Data), are the combined regression model (PR), the fixed effects model (FE), and the random effects model (RE). Special statistical tests were used to choose the best-estimated model, which is the (RE) model. After that, the selected model was subjected to theoretical, statistical, and econometric criteria for analyzing and evaluating the obtained results. It has been shown that each of the determinants—time, trade openness, human development index, corruption control index, and economic growth—has a positive effect on poverty reduction. That is to say, these determinants contribute to poverty reduction, while the results showed that each of the determinants—the Gini coefficient of income disparity, the political instability index, and the globalization index—has a negative effect on the spread of poverty; in other words, they contribute to a rise in the poverty rate of the study sample.The value of the modified coefficient of determination ( ) indicates that about 55% of the changes in poverty for the study sample are attributed to the eight determinants included in the estimated model. Also, the high value of (F) calculated corresponding to a very low probability level (close to zero) indicates that the eight determinants together have a significant effect on changes in poverty. In addition, the quality of reconciliation is very high for the estimated model. Keywords: developing countries, determinants of poverty, panel data, combined regression model (PR), fixed effects model (FE), random effects model (RE).
