Nonconventional random matrix products

Abstract: Let

x i_{1}, x i_{2}, ...

be independent identically distributed random variables and

be a Borel measurable matrix-valued function. Set

X_{n} = F (x i_{q_{1} (n)}, x i_{q_{2} (n)}, ..., x i_{q_{e} l l (n)})

where

0 l e q q_{1} < q_{2} < ... < q_{e} l l

are increasing functions taking on integer values on integers. We study the asymptotic behavior as

N o i n f t y

of the singular values of the random matrix product

P i_{N} = X_{N} c d o t s X_{2} X_{1}

and show, in particular, that (under certain conditions)

f r a c 1 N l o g | P i_{N} |

converges with probability one as

N o i n f t y

. We also obtain similar results for such products when

x i_{i}

form a Markov chain. The essential difference from the usual setting appears since the sequence

(X_{n})

is long-range dependent and nonstationary.

Recommendations

Cites Work

Cited In (4)

This page was built for publication: Nonconventional random matrix products