Time series models for growth of urban population in SAARC countries

The purpose of our present study is to strike out suitable models to explain the growth pattern and to forecast for urban population in SAARC countries. Using the data from UNPD for the years 1950 to 2000 in five years interval, we fitted both exponential and ARMA models. We found the superiority of ARMA models over exponential models to explain the time trend behavior of the urban population as a percentage of total population and to forecast up to the year 2025. We also found that urbanization is faster in Bangladesh than any other SAARC countries. -- urban population ; SAARC countries ; root mean squared forecast error (RMSFE) ; cross validity predictive power (CVPP) ; exponential growth and autoregressive moving average (ARMA)