Abstract
Let be independent and identically distributed random vectors in . Suppose , , where is the identity matrix. Suppose further that there exist positive constants and such that , where denotes the Euclidean norm. Let and let Z be a d-dimensional standard normal random vector. Let Q be a symmetric positive definite matrix whose largest eigenvalue is 1. We prove that for ,
and
where ε and C are positive constants depending only on , and . This is a first extension of Cramér-type moderate deviation to the multivariate setting with a faster convergence rate than . The range of for the relative error to vanish and the dimension requirement for the rate are both optimal. We prove our result using a new change of measure, a two-term Edgeworth expansion for the changed measure, and cancellation by symmetry for terms of the order .
Funding Statement
Fang X. was partially supported by Hong Kong RGC ECS 24301617, GRF 14302418 and 14305821, a CUHK direct grant, and a CUHK start-up grant. Shao Q.M. was partially supported by National Nature Science Foundation of China NSFC 12031005 and Shenzhen Outstanding Talents Training Fund.
Acknowledgments
We thank the two anonymous referees for their careful reading of the manuscript and helpful comments.
Citation
Xiao Fang. Song-Hao Liu. Qi-Man Shao. "Cramér-type moderate deviation for quadratic forms with a fast rate." Bernoulli 29 (3) 2466 - 2491, August 2023. https://doi.org/10.3150/22-BEJ1549
Information