Testing Nonlinear Dynamics, Long Memory and Chaotic Behaviour with Financial and Nonfinancial Data

This paper contains a set of tests for nonlinearities in economic time series. The tests comprise both standard diagnostic tests for revealing nonlinearities and some new developments in modelling nonlinearities. The latter test procedures make use of models in chaos, theory, so-called long-memory models and some asymmetric adjustment models. Empirical tests are carried out with Finnish monthly data for twelve macroeconomic time series covering the period 1920-1996. Test results support unambiguously the notion that there are strong nonlinearities in the data. The evidence for chaos, however, is weak or nonexisting. The evidence on long memory (in terms of so-called rescaled range and fractional differencing) is somewhat stronger although not very compelling. Nonlinearities are detected not only in a univariate setting but also in some preliminary investigation dealing with a multivariate case. Certain differences seem to exist between nominal and real variables in nonlinear behaviour.