The asymptotic Cramer- Rao bound for Gaussian ARMA processes with periodically missing data (Corresp.)

An asymptotic formula is derived for the Cramer-Rao lower bound on unbiased estimates of the parameters of Gaussian autoregressive moving-average (ARMA) processes, in the case where the measurements are not contiguous, but follow a periodic pattern of misses. The formula is then used to illustrate the behavior of the bound for some specific examples.