On the extreme order statistics for stationary Gaussian sequences subject to random missing observations

Abstract: Let $\mathbf{X}={X_{n}}_{n\geq 1}$ be a sequence of stationary Gaussian variables and suppose that only some of the random variables from $\mathbf{X}$ can be observed. In this paper, by studying the limiting properties of multidimensional exceedance point processes for $\mathbf{X}$, we derived the joint limit distribution of extreme order statistics for the Gaussian sequence $\mathbf{X}$ and its observed ones. The joint limit distribution of the locations and heights of the maxima for the Gaussian sequence $\mathbf{X}$ and its observed ones are also obtained.