A nonlinear model for long-memory conditional heteroscedasticity

Abstract: We discuss a class of conditionally heteroscedastic time series models satisfying the equation

r_{t} = z e t a_{t} s i g m a_{t}

, where

z e t a_{t}

are standardized i.i.d. r.v.'s and the conditional standard deviation

s i g m a_{t}

is a nonlinear function

Q

of inhomogeneous linear combination of past values

r_{s}, s < t

with coefficients

b_{j}

. The existence of stationary solution

r_{t}

with finite

p

th moment,

0 < p < i n f t y

is obtained under some conditions on

Q, b_{j}

and

p

th moment of

z e t a_{0}

. Weak dependence properties of

r_{t}

are studied, including the invariance principle for partial sums of Lipschitz functions of

r_{t}

. In the case of quadratic

Q^{2}

, we prove that

r_{t}

can exhibit a leverage effect and long memory, in the sense that the squared process

r_{t}^{2}

has long memory autocorrelation and its normalized partial sums process converges to a fractional Brownian motion.

Recommendations

Cites work

Cited in

(11)

This page was built for publication: A nonlinear model for long-memory conditional heteroscedasticity