<![CDATA[The use of quantile regression in analyzing determinants of life expectancy provides a deep understanding of the influence of predictor variables across various quantiles of the data distribution. The analysis results reveal significant differences in predictor variables affecting life expectancy at each quantile, with socio-economic, environmental, and health program factors having varying impacts. Model evaluation is conducted considering several goodness-of-fit measures such as MAPE, AIC, and coefficient of determination, selecting the best quantile based on a combination of these values. Based on the analysis, the optimal quantile is chosen based on the lowest MAPE and AIC, and the highest coefficient of determination, considering the efficiency of significant predictor variables. The analysis indicates that the best model is found at the 60th quantile, with two significant predictor variables, namely mean years of schooling and labor force participation. This underscores the significant role of socio-economic factors in determining life expectancy at specific quantiles]]>
Copyrights © 2024
