We consider high-frequency sampled continuous-time autoregressive moving average (CARMA) models driven by finite-variance zero-mean Lévy processes. An L²-consistent estimator for the increments of the driving Lévy process without order selection in advance is proposed if the CARMA model is invertible. In the second part, the underlying process of the CARMA model is chosen to be either a symmetric alpha-stable Lévy process or a symmetric Lévy process with finite second moments. In the doubly asymptotic framework of high-frequency data within a long time interval, convergence of normalized and self-normalized versions of the periodogram to functions of stable distributions is shown. Moreover, a consistent estimate for the normalized power transfer function is established by applying a smoothing filter to the periodogram. This result is used to propose an estimator for the parameters of the CARMA process.
«
We consider high-frequency sampled continuous-time autoregressive moving average (CARMA) models driven by finite-variance zero-mean Lévy processes. An L²-consistent estimator for the increments of the driving Lévy process without order selection in advance is proposed if the CARMA model is invertible. In the second part, the underlying process of the CARMA model is chosen to be either a symmetric alpha-stable Lévy process or a symmetric Lévy process with finite second moments. In the doubly asym...
»