Modeling Trends in the Variables of the Money Demand Function
As discussed in the previous chapter the issue of whether economic time series are nonstationary or not is important for both estimation and hypothesis testing, both of which rely on asymptotic distribution theory. Moreover, the nature of nonstationarity has important implications for the appropriate transformation to attain a stationary series as well as for the estimation of long-run relationships between nonstationary variables.
KeywordsLyapunov Exponent Unit Root Unit Root Test Money Demand Financial Time Series
Unable to display preview. Download preview PDF.
- 4.The idea here is that the choice of a break point should be an explicit part of the estimation procedure, because in practice one never selects a date to test for a break point without prior information about the data. Moreover, endogenizing the break point leads to critical values that are much more conservative that Perron’s (1989) ones.Google Scholar
- 5.See Barnett and Serletis (2000) for several other univariate statistical tests for independence, nonlinearity and chaos, that have been recently motivated by the mathematics of deterministic nonlinear dynamical systems.Google Scholar
- 6.Another very promising approach to the estimation of Lyapunov exponents [that is similar in some respects to the Nychka et ai (1992) approach] has also been recently proposed by Ramazan Gencay and Davis Dechert (1992). This involves estimating all Lyapunov exponents of an unknown dynamical system. The estimation is carried out, as in Nychka et al. (1992), by a multivariate feedforward network estimation technique — see Gencay and Dechert (1992) for more details.Google Scholar
- 7.See Serletis (1995) or Serletis and Ioannis Andreadis (2000) for more details regarding these issues.Google Scholar