On the distributions of \(L_ p\) norms of weighted quantile processes (Q913410)

Let \(X_ 1,X_ 2,..\). be i.i.d. r.v.'s with distribution F and density f. Define the quantile function by \(Q(t)=\inf \{x:\) F(x)\(\geq t\}\) \((0<t<1)\), and the empirical quantile function by \(Q_ n(t)=X_{k,n}\) if \((k-1)/n<t\leq k/n\) where \(X_{1,n}\leq X_{2,n}\leq...\leq X_{n,n}\) are the order statistics of \(X_ 1,X_ 2,...,X_ n\). The quantile process is defined as \[ \rho_ n(t)=n^{1/2}f(Q(t))(Q(t)-Q_ n(t)). \] One of the main results claims that \(\int^{1/2}_{1/(n+1)}| \rho_ n(t)|^ p/q(t)dt\) converges in distribution to \(\int^{1/2}_{0}| B(t)|^ p/q(t)dt\) as \(n\to \infty\) where \(1\leq p<\infty\), B(t) is the Brownian bridge and q(t) is a positive- valued function satisfying some regularity conditions.