In this note, the Bahadur representation of sample quantile is considered in the case where observations satisfy some strongly mixing condition. The results in this note extend Sen's results which are obtained in some ø-mixing cases.

Let {[var epsilon]n1,...,[var epsilon]nn;n[greater-or-equal, slanted]1} be a sequence of series of random variables that are independently and identically distributed within each series. PutSn,i=[var epsilon]n1+...+[var epsilon]ni. We prove that under the conditions which assure the validity of...

We consider the asymptotic normality of the random variable Z([lambda]) = [Sigma]cj([lambda])[zeta]jas[lambda] --> [infinity] where {[zeta]j} is a strictly stationary strongly mixing sequence.