Nonlinear Correlograms and Partial Autocorrelograms*
- 6 December 2005
- journal article
- Published by Wiley in Oxford Bulletin of Economics and Statistics
- Vol. 67 (s1) , 957-982
Abstract
This paper proposes neural network‐based measures of predictability in conditional mean, and then uses them to construct nonlinear analogues to autocorrelograms and partial autocorrelograms. In contrast to other measures of nonlinear dependence that rely on nonparametric estimation of densities or multivariate integration, our autocorrelograms are simple to calculate and appear to work well in relatively small samples.Keywords
All Related Versions
This publication has 36 references indexed in Scilit:
- DIAGNOSTIC CHECKING FOR THE ADEQUACY OF NONLINEAR TIME SERIES MODELSEconometric Theory, 2003
- Optimal Forecast Combinations Under General Loss Functions and Forecast Error DistributionsSSRN Electronic Journal, 2002
- Nonlinear Autocorrelograms: an Application to Inter‐Trade DurationsJournal of Time Series Analysis, 2002
- A test for independence based on the correlation dimensionEconometric Reviews, 1996
- Specification, Estimation, and Evaluation of Smooth Transition Autoregressive ModelsJournal of the American Statistical Association, 1994
- POWER OF THE NEURAL NETWORK LINEARITY TESTJournal of Time Series Analysis, 1993
- Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networksNeural Networks, 1990
- Testing and Modeling Threshold Autoregressive ProcessesJournal of the American Statistical Association, 1989
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978
- A new look at the statistical model identificationIEEE Transactions on Automatic Control, 1974