Peng  proposed an asymptotically normal estimator of the mean of a heavy tailed distribution with tail index a > 1 based on an i.i.d. observations. The goal in this paper is to propose an extension of this estimator which is also asymptotically normal for a sequence X1,X2, . . . ,Xn, . . . resulting from an AR(1) stationary process with common heavy tailed distribution of innovations.
Fawzi Mami, Tawfiq; Ouadjed, Hakim; and Yousfate, Abderrahmane
"Estimating the Mean of an AR(1) Process with Infinite Variance,"
Journal of Statistics Applications & Probability: Vol. 4
, Article 19.
Available at: https://dc.naturalspublishing.com/jsap/vol4/iss1/19