• Title/Summary/Keyword: B-valued random variables

Search Result 4, Processing Time 0.015 seconds

On the Probability Inequalities under Linearly Negatively Quadrant Dependent Condition

  • Baek, Jong Il;Choi, In Bong;Lee, Seung Woo
    • Communications for Statistical Applications and Methods
    • /
    • v.10 no.2
    • /
    • pp.545-552
    • /
    • 2003
  • Let X$_1$, X$_2$, … be real valued random variables under linearly negatively quadrant dependent (LNQD). In this paper, we discuss the probability inequality of ennett(1962) and Hoeffding(1963) under some suitable random variables. These results are to extend Theorem A and B to LNQD random variables. Furthermore, let ζdenote the pth quantile of the marginal distribution function of the $X_i$'s which is estimated by a smooth estima te $ζ_{pn}$, on the basis of X$_1$, X$_2$, …$X_n$. We establish a convergence of $ζ_{pn}$, under Hoeffding-type probability inequality of LNQD.

Complete Convergence in a Banach Space (바나하 공간에서의 완전 수렴성)

  • Sung, Soo-Hak
    • The Journal of Natural Sciences
    • /
    • v.9 no.1
    • /
    • pp.57-60
    • /
    • 1997
  • Let {$X_{ni}$,1$\leq$i$\leq$,n$\geq$1} be an array of rowwise independent B-valued random variables which is uniformly bounded by a random various X satisfying $E|X|^{2p}<\infty$ for some p$\geq$1. Let {$a_{ni}$,1$\leq$i$\leq$,n$\geq$1} be an array of constants. Under some auxiliary conditions on {$a_{ni}$}, it is shown that $sum_{i=1}^n a_{ni}X_{ni}\rightarrow0$ in probability if and only if $sum_{i=1}^n a_{ni}X_{ni}$ converges completely ot 0.

  • PDF

A Note on a Result of Yu. V. Prokhorov in General Banach Spaces

  • Dug Hun Hong
    • Communications for Statistical Applications and Methods
    • /
    • v.4 no.1
    • /
    • pp.255-258
    • /
    • 1997
  • We prove a conjecture of Yu. V. Prokhorov in general Banach Spaces ; let ($X_n$, n$\geq$1} be a sequence of independent identically and symmetrically distributed Banach valued random variables, then the relation $\mid$$\mid$$S_n$$\mid$$\mid$/$b_n$ -> 1 a.s. cannot hold for any choice of constants $b_n$.

  • PDF