Title: Bayesian smoothing spline model and its application in current population survey
Authors: Zhuoqiong He - University of Missouri (United States) [presenting]
Abstract: The Current Population Survey (CPS) is conducted to collect the labor force data and to measure the extent of unemployment in the United States of America. The total numbers of employment population and unemployment population are estimated from the CPS sample, and seasonal adjustment of the estimates is needed to observe the changing of economic conditions. We propose a Bayesian smoothing spline (BSS) model to remove the seasonal fluctuations and to capture the fundamental tendency of a labor force total associated with general economic expansions and contractions. This BSS model can be efficiently computed with Markov chain Monte Carlo. The estimation of unemployment based on BSS is illustrated and compared with the seasonally adjusted unemployment estimation published by U.S. bureau of labor statistics.