Reconsideration of the Markov Chain Evidence on Unemployment Rate Asymmetry

Using 15 years worth of additional data, a study is carried out to explore the extent to which the results in Rothman (1991) still hold. Using raw unfiltered data, the aggregate unemployment rate appears to be a Neftci-type symmetric process. But use of two detrending procedures produces strong evidence in favor of asymmetry for this series. The most robust result is that the manufacturing sector unemployment rate is strongly asymmetric across use of both the unfiltered and detrended data. The analysis also shows that reliance on the asymptotic distribution of the underlying test statistics can often lead to spurious inference. JEL Categories: C22, E32 ∗Department of Economics, Brewster Building, East Carolina University, Greenville, NC 27858 USA, email: [email protected], phone: 252-328-6151. Support from an East Carolina University Thomas Harriot College of Arts and Sciences research grant is gratefully acknowledged.