Exploring New Models for Population Prediction in Detecting Demographic Phase Change

by Arindam Gupta, Sabyasachi Bhattacharya ,Asis Kumar Chattyopadhyay.

Abstract: Logistic model has long history regarding it's usefulness in population predictions. But the model has some limitations when applying for the sparse census data sets, typically available for developing countries. In such situation the relative growth rates (RGR) exhibit some unusual trend (increasing, primary increasing and then decreasing) which is different from the common decreasing trend of logistic law. To tackle those complicated demographic situations we have successfully explored a simplified version of Tsoularis and Wallace model (TWM) which can explain all of these feasible monotonic structures of RGR. In addition to this we have also proposed another model (PM) by assuming RGR as a direct function of time covariate but not the size. The model has some key advantages than the simplified TWM (STWM). It can detect the demographic phase change point at which the developing country switches over towards developed one. We performed RGR modelling (as a function of time) but not the size as neither TWM nor STWM is analytically solvable and the underlying population model is better identifiable in the former case but not in the later. The less number of parameters involve in both the STWM and PM ensure a better chance of convergence under non-linear least square estimation than the original TWM with more parameters.

Key Words: Population prediction, relative growth rate, Tsoularis-Wallace model, demographic phase change

Arindam Gupta, arindamdeep@yahoo.com
Sabyasachi Bhattacharya, bsabya2000@yahoo.co.in
Asis Kumar Chattyopadhyay, asis_stat@yahoo.com

Editor: Kishore K Das,daskkishore@gmail.com

READING THE ARTICLE: You can read the article in portable document (.pdf) format (606822 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2490 times since October 17, 2007.

Return to the InterStat Home Page.