Modeling and forecasting population growth rate in Bangladesh: Using Arima models
Md Muhaiminul Islam Selim, Md Ahasan Ul Haque, Liton Chandra Voumik
Background/Objectives: The tremendous increase in population is one of the most crucial problems of Bangladesh. Currently, one of the most densely populated countries and this bulky size of the population is now one of the main concerns in this country. This paper aims to apply the ARIMA method to model and forecast population growth rates in Bangladesh. Methods/ Statistical analysis: The data was collected from World Development Indicators (WDI) and it has been collected over 40 years by WDI, World Bank. We applied Phillips–Perron (PP) and Augmented Dickey-Fuller (ADF) tests to investigate the stationary character of the data. Stata and R statistical software was used to build a structure of the ARIMA method to model and forecast the population growth rates. Findings: In this study, the population of Bangladesh from 1979 to 2018 is modeled using ARIMA (P, I, Q) methodology. The model was validated by the lowest values of AIC and BIC, less than 5% of P-values, graphical presentations of ACF and PACF plots, and corrgrams. Using these models, the numeric figure of future population growths are drawn and forecasted. The population growth rate for the next first and second decades is also forecasted using the ARIMA (1, 1, 1) model. If the decreasing trend persists, the population growth rate of Bangladesh would be approximately 0.56% in 2028 and 0.35% in 2038. Statistical outcomes illustrate that Bangladesh’s population growth rate is a decreasing trend that will continue declining in the future. Improvements/Applications: This finding will help policymakers and academicians to formulate population-related strategies and policies more precisely.