Nuisance parameter free properties of correlation integral based statistics
- 1 January 1996
- journal article
- research article
- Published by Taylor & Francis in Econometric Reviews
- Vol. 15 (3) , 237-259
- https://doi.org/10.1080/07474939608800354
Abstract
This paper presents conditions under which statistics based on the correlation integral are invariant to the use of estimated residuals from dynamic models. The methods are applied to the so-called BDS test for nonlinearity, under a variety of data-generating processes.Keywords
This publication has 22 references indexed in Scilit:
- Consistency and Limiting Distribution of the Least Squares Estimator of a Threshold Autoregressive ModelThe Annals of Statistics, 1993
- Quasi-maximum likelihood estimation and inference in dynamic models with time-varying covariancesEconometric Reviews, 1992
- GARCH(1, 1) processes are near epoch dependentEconomics Letters, 1991
- Time Series: Theory and MethodsPublished by Springer Nature ,1991
- Ergodic theory of chaos and strange attractorsReviews of Modern Physics, 1985
- Non-strong mixing autoregressive processesJournal of Applied Probability, 1984
- Measuring the strangeness of strange attractorsPhysica D: Nonlinear Phenomena, 1983
- On the Strong Mixing Property for Linear SequencesTheory of Probability and Its Applications, 1978
- Strong mixing properties of linear stochastic processesJournal of Applied Probability, 1974
- Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series ModelsJournal of the American Statistical Association, 1970