One of the dimensions of the Sustainable Development Goals (SDGs) is the economy. The performance of the economy can be seen from macroeconomic indicators, namely Gross Domestic Product (GDP), inflation, labor force index, and interest rates. The SDGs are also very closely related to demographic conditions, especially the level of health and welfare which is reflected in life expectancy. In this study, it is proposed to model life expectancy with macroeconomic variables using the Neural Network (NN) method. The selection of the NN method is based on the existence of a non-linear relationship between the life expectancy variable and several macroeconomic variables. The data used is Indonesia's annual data for the period 1990-2020. NN modeling uses several combinations, namely a combination of inputs, number of neurons, and activation functions. Based on these combinations, the best model is obtained, namely the NN(4,4,1) model with the activation function of tanh. The model explains that the four macroeconomic variables have an effect on life expectancy. This is in accordance with the argument in the book Macroeconomics that life expectancy modeling can be influenced by macroeconomic variables.
Copyrights © 2022