We study the joint limit distribution of the largest eigenvalues of a sample covariance matrix based on a large matrix . The rows of are given by independent copies of a linear process, , with regularly varying noise with tail index . It is shown that a point process based on the eigenvalues of converges, as and at a suitable rate, in distribution to a Poisson point process with an intensity measure depending on and . This result is extended to random coefficient models where the coefficients of the linear processes are given by , for some ergodic sequence , and thus vary in each row of . As a by-product of our techniques we obtain a proof of the corresponding result for matrices with iid entries in cases where goes to zero or infinity and .
View on arXiv