Life Expectancy Prediction Based on GDP and Population Size of Bangladesh using Multiple Linear Regression and ANN Model

Life Expectancy is one of the most basic significant aspects of natives that can have a significant relationship between the health sector and the financial movement of a country. This paper investigates the correlation between the Gross Domestic Product (GDP) and Population with Life Expectancy (LE) of Bangladesh. To achieve our goal, we used a long period of data from World Data Open Data (WBOD) and Trends Economics from 1960 to 2020. Using a different number of prediction models was applied as a Multiple Linear Regression Model and several Artificial Neural Network (ANN) to evaluate the life expectancy of Bangladeshis people. There are two independent variables (GDP & Population) used to assess the dependent variable as life expectancy. Our applied model indicates GDP can impact life expectancy, and it refers to longer life expectancy for Bangladeshi's people. There is a strong correlation between population size with life expectancy. Among the applied model, the Multiple Linear Regression (MLR) model stands at the top position with 98% accuracy, and another model, the multi-feed forwarded artificial neural network (MFFNN), regarding the combination of (8,1) with two hidden layer reaches 94% accuracy. Our examination alludes to the life expectancy from 2009 to 2026 of Bangladeshis people. This study shows that GDP improves life expectancy and that population has an impact on life expectancy, and it recommended that the research be expanded with more data and machine learning algorithms.