A General Dependence Test and Applications

David S. Johnson and Robert McClelland

Abstract

We describe a test, based on the correlation integral, for the independence of a variable and a vector that can be used to detect model misspecification in serially dependent data. In Monte Carlo simulations this test performs nearly as well or better than the BDS test in univariate time series and complements the BDS test in distributed lag models. Finally, we apply our test to detect misspecification in models of U.S. unemployment data.