Second Moment Estimator for An AR(1) Model Driven by A Long Memory Gaussian Noise
Author:
Preprint
Abstract
In this paper, we consider an inference problem for the first order autoregressive process driven by a long memory Gaussian process. The assumptions and results are stated in terms of the covariance function of the noise. Some examples include the fractional Gaussian noise and the fractional ARIMA model. For the second moment estimator, we prove the strong consistency and the asymptotic normality, and obtain the Berry-Esseen bound under appropriate conditions. The proof is based on the decay rate of the stationary solution's covariance associated with the Gaussian noise.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Parameter estimation for the Langevin equation with stationary-increment Gaussian noiseStatistical Inference for Stochastic Processes, 2017
- ASYMPTOTIC INFERENCE FOR AR MODELS WITH HEAVY-TAILED G-GARCH NOISESEconometric Theory, 2014
- Large Sample Inference for Long Memory Processes2012
- The Pathwise Convergence of Approximation Schemes for Stochastic Differential EquationsLMS Journal of Computation and Mathematics, 2007
- Asymptotic normality of regression estimators with long memory errorsStatistics & Probability Letters, 1996
- Asymptotic Properties of the LSE in a Regression Model with Long-Memory Stationary ErrorsThe Annals of Statistics, 1991
- Central limit theorems for non-linear functionals of Gaussian fieldsJournal of Multivariate Analysis, 1983
- Asymptotic properties of general autoregressive models and strong consistency of least-squares estimates of their parametersJournal of Multivariate Analysis, 1983
- Strong Consistency of Least Squares Estimates in Dynamic ModelsThe Annals of Statistics, 1979
- Convergence of integrated processes of arbitrary Hermite rankProbability Theory and Related Fields, 1979