A Non-Parametric Approach to Extrapolate the Demographic Indicators of Small Subsections of the Population

Projections of the demographic indicators at the national and state level have been made for many years but for decentralized planning and better monitoring of already existing policies projections are needed for smaller subsections of the population. In this paper, an attempt has been made to present a simple nonparametric technique to extrapolate the demographic indicator Total Fertility Rate (TFR) at granular levels to a recent future, where the projected value of that indicator is available at the gross level only by applying a data mining technique. The approach has been used to project the total fertility rate of women in the districts of Uttar Pradesh, India for the year 2026. From this study, the TFR of Uttar Pradesh is found to be 1.98 for the year 2026 which is below the replacement level of fertility. But still, there are 28% (20 out of 71 districts) of the districts have TFR above the replacement level of fertility.. KEYWORDS :Projection, TFR, Non-parametric, Granular level, Replacement level of fertility.