|
Sufficient conditions for the Marchenko–Pastur theorem
P. A. Yaskov Steklov Mathematical Institute of Russian Academy of Sciences, Moscow
Abstract:
We find general sufficient conditions in the Marchenko–Pastur theorem
for high-dimensional sample covariance matrices associated with random vectors, for which the weak concentration property of quadratic forms may not hold in general. The results are obtained by means of a new martingale method, which may be useful in other problems of random matrix theory.
Keywords:
random matrices, sample covariance matrices, the Marchenko–Pastur law.
Received: 26.06.2023 Accepted: 18.09.2023
Citation:
P. A. Yaskov, “Sufficient conditions for the Marchenko–Pastur theorem”, Teor. Veroyatnost. i Primenen., 68:4 (2023), 813–833; Theory Probab. Appl., 68:4 (2024), 657–673
Linking options:
https://www.mathnet.ru/eng/tvp5668https://doi.org/10.4213/tvp5668 https://www.mathnet.ru/eng/tvp/v68/i4/p813
|
Statistics & downloads: |
Abstract page: | 157 | Full-text PDF : | 12 | References: | 34 | First page: | 10 |
|