Fluctuations of the empirical quantiles of independent Brownian motions

Abstract: We consider

n

independent, identically distributed one-dimensional Brownian motions,

B_{j} (t)

, where

B_{j} (0)

has a rapidly decreasing, smooth density function

f

. The empirical quantiles, or pointwise order statistics, are denoted by

B_{j : n} (t)

, and we are interested in a sequence of quantiles

Q_{n} (t) = B_{j (n) : n} (t)

, where

j (n) / n o a l p h a i n (0, 1)

. This sequence converges in probability in

C [0, i n f t y)

to

q (t)

, the

a l p h a

-quantile of the law of

B_{j} (t)

. Our main result establishes the convergence in law in

C [0, i n f t y)

of the fluctuation processes

F_{n} = n^{1 / 2} (Q_{n} - q)

. The limit process

F

is a centered Gaussian process and we derive an explicit formula for its covariance function. We also show that

F

has many of the same local properties as

B^{1 / 4}

, the fractional Brownian motion with Hurst parameter

H = 1 / 4

. For example, it is a quartic variation process, it has H"older continuous paths with any exponent

g a m m a < 1 / 4

, and (at least locally) it has increments whose correlation is negative and of the same order of magnitude as those of

B^{1 / 4}

.

Recommendations

Cites Work

Cited In (9)

This page was built for publication: Fluctuations of the empirical quantiles of independent Brownian motions