Rosenthal-type inequalities for linear statistics of Markov chains
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909805393739776 |
|---|---|
| author | Durmus, Alain Moulines, Eric Naumov, Alexey Samsonov, Sergey Sheshukova, Marina |
| author_facet | Durmus, Alain Moulines, Eric Naumov, Alexey Samsonov, Sergey Sheshukova, Marina |
| contents | In this paper, we establish novel concentration inequalities for additive functionals of geometrically ergodic Markov chains similar to Rosenthal inequalities for sums of independent random variables. We pay special attention to the dependence of our bounds on the mixing time of the corresponding chain. Precisely, we establish explicit bounds that are linked to the constants from the martingale version of the Rosenthal inequality, as well as the constants that characterize the mixing properties of the underlying Markov kernel. Finally, our proof technique is, up to our knowledge, new and is based on a recurrent application of the Poisson decomposition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_05838 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Rosenthal-type inequalities for linear statistics of Markov chains Durmus, Alain Moulines, Eric Naumov, Alexey Samsonov, Sergey Sheshukova, Marina Probability Statistics Theory Machine Learning 60E15, 60J20, 65C40 In this paper, we establish novel concentration inequalities for additive functionals of geometrically ergodic Markov chains similar to Rosenthal inequalities for sums of independent random variables. We pay special attention to the dependence of our bounds on the mixing time of the corresponding chain. Precisely, we establish explicit bounds that are linked to the constants from the martingale version of the Rosenthal inequality, as well as the constants that characterize the mixing properties of the underlying Markov kernel. Finally, our proof technique is, up to our knowledge, new and is based on a recurrent application of the Poisson decomposition. |
| title | Rosenthal-type inequalities for linear statistics of Markov chains |
| topic | Probability Statistics Theory Machine Learning 60E15, 60J20, 65C40 |
| url | https://arxiv.org/abs/2303.05838 |