A Statistical Approach to a Dynamic Flow Model of the Labor Market

• Activity Number: 417
• Type: Contributed
• Date/Time: Tuesday, August 2, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Mingfei Li*+ and Mihaela Predescu
• Companies: Bentley University and Bentley University
• Keywords: Labor market ; business statistics ; nonlinear system ; parameter estimation

Abstract

Neugart (2002) proposed a dynamic flow model of the labor market with nonlinear and endogenous outflow rate from unemployment. Inflation as a feedback on job offers through the real money supply was also involved in this model system. And in this unemployment and inflation deterministic model system, it is shown that the stability of "equilibrium of rate of unemployment" affected by the parameters of this system. However, how these parameters should be determined in this system has not been discussed. In this talk we would like to show that for practical purpose, determination of the parameters in this dynamic model is critical, i.e. bad choice of parameters may lead the whole model fail in describing or forecasting unemployment and inflation. We start our stochastic approach from simply including random error terms in the deterministic model, and apply MLE method on the parameters of the system. Simulations on real labor market will show the performance of the model with statistical estimated parameters in predicting real labor market. Furthermore, we discuss an adjusted model by release endogenous inflow rate of labor market.