Regression Model Selection for Life Expectancy Prediction: A Comparative Analysis of Imputation Techniques

This study carried out a comparative analysis of the mean, median, and mode imputations for NULL values in a dataset. We used five different pre-processing techniques to create three distinct regression models, including tree-based models like decision trees and random forests, as well as linear and support vector regression. This allows us to draw different conclusions. The main objective of this study is to compare these findings and arrive at the reasons behind the output of different imputation techniques. The regression model selection helps to explain these results. Using various qualitative measures, the findings of the proposed method were tested and verified for lifetime prediction efficiency and performance The study showed that the random forest regression method achieved accuracy the highest being 96.8%. These results emphasize the importance of random forests in comparison to alternative regression methods used innovatively for life expectancy estimation.