Nonparametric regression for locally stationary time series
arXiv:1302.4198 · doi:10.1214/12-AOS1043
Abstract
In this paper, we study nonparametric models allowing for locally stationary regressors and a regression function that changes smoothly over time. These models are a natural extension of time series models with time-varying coefficients. We introduce a kernel-based method to estimate the time-varying regression function and provide asymptotic theory for our estimates. Moreover, we show that the main conditions of the theory are satisfied for a large class of nonlinear autoregressive processes with a time-varying regression function. Finally, we examine structured models where the regression function splits up into time-varying additive components. As will be seen, estimation in these models does not suffer from the curse of dimensionality.
Published in at http://dx.doi.org/10.1214/12-AOS1043 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
References in corpus (6)
- Statistical inference for time-varying ARCH processes
- Local linear quantile estimation for nonstationary time series
- Mixing properties of ARCH and time-varying ARCH processes
- Normalized least-squares estimation in time-varying ARCH models
- On recursive estimation for time varying autoregressive processes
- Nonparametric inference of quantile curves for nonstationary time series