USING A COMPOSED ERROR MODEL TO ESTIMATE THE FRICTIONAL AND EXCESS‐SUPPLY COMPONENTS OF UNEMPLOYMENT
提出一种随机模型，将误差项分为双边和单边两部分，用于估计失业的摩擦性和超额供给成分，并分析美国各州摩擦性失业率的经济和人口决定因素。
ABSTRACT We present in this paper a plausible and simple method of estimating the two components (frictional and excess supply) of unemployment. This approach uses a stochastic model whose error term is composed of two elements–the usual two‐sided error and a one‐sided error. Our method has several strengths. First, we are able to explicitly model the universally held view that there is a nonzero lower bound on unemployment. Second, we can easily determine whether each region's unemployment rate is caused primarily by excess supply or frictional forces. We illustrate our technique on a data set comprised of all 50 states over the period 1960‐1979. Finally, estimation of the frictional rates of unemployment allows us to analyze, in the last part of the paper, the underlying economic and demographic determinants of differences in frictional unemployment rates across states and over time.