Population Forecasting by Population Growth Models based on MATLAB Simulation

Abstract

Listen

Translate article

Population forecasting is a necessary effort to understand population growth, which affects various aspects of a country’s society and economy, including future demand for food, water, energy, and services. Mathematical models are commonly used to understand the interplay of the migration, birth and death rates on population growth. Mathematical models help population forecasting by capturing statistical trends from historical datasets. However, these need to be carefully compared to understand the implications of different model formulations in predicting future population, which the models have not seen or were trained on. In this work, we have compared two common population growth models, namely Malthusian law and Logistic law, using US population data from 1951 to 2019. By formulating a least-squared curve fitting problem, the birth and death rates can be estimated using MATLAB software. MATLAB simulations showed that the Logistic law of population growth yields smaller sum of squared residuals than the Malthusian law. In this case, a better population model may be beneficial in the social science, such as political science and sociology.