On asymptotic minimaxity of fixed accuracy estimators for autoregression parameters. I: Stable process (Q1815811)

The present paper deals with estimation of parameters of autoregression processes and its object is to establish the uniform asymptotic normality and the asymptotic optimality in the minimax sense of the fixed accuracy estimators proposed by the authors in Avtom. Telemekh 1988, No. 11, 130-141 (1988); English translation in Autom. Remote Control 49, No. 11, 1495-1504 (1988; Zbl 0699.93089). These properties are studied separately for stable and explosive autoregression processes (the latter case to be treated in the forthcoming Part II) because these processes have a completely different statistical nature.