1: \begin{abstract}

2: Interest in continuous-time processes has increased rapidly in recent years, largely because of high-frequency data available in many applications.  We develop a method for estimating the kernel function $g$ of a second-order stationary L\'evy-driven continuous-time moving average (CMA) process $Y$ based on observations of the discrete-time process $Y^\Delta$ obtained by sampling $Y$ at $\Delta, 2\Delta,\ldots,n\Delta$ for small $\Delta$.  We approximate $g$ by $g^\Delta$ based on the Wold representation  and prove its pointwise convergence to $g$ as $\Delta\rightarrow 0$ for $\CARMA(p,q)$ processes.  Two non-parametric estimators of $g^\Delta$, based on the innovations algorithm and the Durbin-Levinson algorithm, are proposed to estimate $g$. For a Gaussian CARMA process we give conditions on the sample size $n$ and the grid-spacing $\Delta(n)$ under which the innovations estimator is consistent and asymptotically normal as $n\rightarrow\infty$.

3: The estimators can be calculated from sampled observations of {\it any} CMA process and simulations suggest that they perform well even outside the class of CARMA processes.  

4: We illustrate their performance for simulated data

5: and apply them to the Brookhaven turbulent wind speed data.  

6: Finally we extend results of \citet{bfk:2011:1} for sampled CARMA processes to a much wider class of CMA processes.

7: \end{abstract}

8: