Abstract
© 2018 Elsevier B.V. Given any stationary time series {Xn:n∈Z} satisfying an ARMA(p,q) model for arbitrary p and q with infinitely divisible innovations, we construct a continuous time stationary process {xt:t∈R} such that the distribution of {xn:n∈Z}, the process sampled at discrete time, coincides with the distribution of {Xn}. In particular the autocovariance function of {xt} interpolates that of {Xn}.
Original language | English |
---|---|
Pages (from-to) | 156-167 |
Journal | Journal of Statistical Planning and Inference |
Volume | 197 |
DOIs | |
Publication status | Published - 1 Dec 2018 |
Keywords
- CARMA
- Continuous-time ARMA
- Discrete-time ARMA
- Embedding
- Lévy process