Showing papers by "Melvin J. Hinich published in 1998"

Episodic nonstationarity in exchange rates

Abstract: We examine a method recently proposed by Hinich and Patterson (mimeo, University of Texas at Austin, 1995) for testing the validity of specifying a GARCH error structure for financial time series data in the context of a set of ten daily Sterling exchange rates. The results demonstrate that there are statistical structures present in the data that cannot be captured by a GARCH model, or any of its variants. This result has important implications for the interpretation of the recent voluminous literature which attempts to model financial asset returns using this family of models.

Frequency-domain test of time reversibility

Abstract: We introduce a frequency-domain test of time reversibility, the REVERSE test. It is based on the bispectrum. We analytically establish the asymptotic distribution of the test and also explore its finite-sample properties through Monte-Carlo simulation. Following other researchers who demonstrated that the problem of business-cycle asymmetry can be stated as whether macroeconomic fluctuations are time irreversible, we use the REVERSE test as a frequency-domain test of business-cycle asymmetry. Our empirical results show that time irreversibility is the rule rather than the exception for a representative set of macroeconomic time series for five OECD countries.

Ideology and the construction of nationality: The Canadian elections of 1993

Abstract: Canada is one nation, but it is in many ways two communities, one Francophone and the other Anglophone. We employ a formal model of “ideology” and analyze how nationality is constructed in people's minds. The magnitude of the changes in expressed “preferences” in terms of ideology depends on the salience of the new issue, the extent to which it confirms with the existing ideological cleavage, and the difference between the perceived status quo on the new dimension and the voter's most preferred alternative. Using data from the 1993 Canadian National Election Study, we consider the relative importance of different policy dimensions in explaining voting decisions among educated Canadians. The issue of Quebec sovereignty, alone, is shown to have significant power for predicting vote choice. A plausible explanation, confirmed here by regression analysis, is that Quebec sovereignty “stands” for other issues in voters' conception of Canadian politics.

Forecasting High Frequency Exchange Rates Using Cross-Bicorrelations

Abstract: The central motivation for this paper is to combine a multivariate nonlinearity test with a nonlinear time series forecasting model of the VAR-type which incorporates lags of the cross-products with the other exchange rate in the system using high frequency data. The paper draws together two sets of literature which have previously been considered entirely separately: namely time series tests for nonlinearity and multivariate exchange rate forecasting. The out-of-sample forecasting performance of the new technique using cross-bicorrelations is inferior to that of univariate time series models, in spite of significant in-sample cross-bicorrelation statistics, although it does seem to be able to predict the signs of the returns well in certain cases. A number of explanations for this apparent paradox are proposed.

Nonlinear changes in evoked potentials during recovery

Abstract: Evoked potential (EP) recordings remain a sensitive indicator of the cerebral response to episodes of brain injury. Previous investigations focused primarily on power and spectral (Fourier) indicators of injury. These parameters are derived from second order statistics. Nonlinear changes also occur after the induction of hypoxic- ischemic (HI) injury. These alterations may be captured using higher order statistics such as the bicorrelation or third order cumulants. Higher order statistics are able to accurately capture signal shape changes due to phase alterations. The H-statistic examines the size of the lagged terms of the third order cumulant. Large H-statistics reveal embedded quadratic nonlinearities. Additionally, this statistic operates on single evoked potentials. In an animal model of HI injury we show that the baseline evoked potential morphology maintains a high degree of nonlinearity. More than one third of these baseline cases have high H-statistics (p>.95). After recovery from HI injury, the H-statistic shows a uniform distribution where an expected 5% of the cases adopt this probability value. H- statistic changes reveal a distinctive loss of nonlinear character after HI injury.