Analysis on Relevant Factors Affecting Life Expectancy

Although human life expectancy has exceeded 70 years, life expectancy in some countries is still short. Studying the factors affecting life expectancy is significant to improve life expectancy. Therefore, in this work, we firstly analyzed the factors related to life expectancy and then designed the multiple linear regression (MLR) models to predict life expectancy. In this work, 18 factors were considered in regression model, such as GDP, disease, school index, etc. According to Spearman correlation coefficient, we analyzed the reasons affecting the life expectancy, and design 3 MLR models to predict life expectancy. The result show that disease, schooling index and income are considerable factors affecting life expectancy. The root mean squared error (RMSE) of MLR is 3.818 obtained after analysis. In addition, we verified that factors affecting life expectancy are different for developed and developing countries. So, it is necessary to regress for two types of countries. For developing countries, medical and school factors are more important for developing countries. The RMSE of MLR is 3.99 for developing countries, and 2.57 for developed countries.