Exact Moderate and Large Deviations for Linear Processes

  title={Exact Moderate and Large Deviations for Linear Processes},
  author={Magda Peligrad and Hailin Sang and Yu Yong Zhong and Wei Biao Wu},
  journal={arXiv: Statistics Theory},
Large and moderate deviation probabilities play an important role in many applied areas, such as insurance and risk analysis. This paper studies the exact moderate and large deviation asymptotics in non-logarithmic form for linear processes with independent innovations. The linear processes we analyze are general and therefore they include the long memory case. We give an asymptotic representation for probability of the tail of the normalized sums and specify the zones in which it can be… 

Figures from this paper

Moderate Deviations for Interacting Processes
This article is concerned with moderate deviation principles of a class of interacting empirical processes. We derive an explicit description of the rate function, and we illustrate these results
Performance bounds for parameter estimates of high-dimensional linear models with correlated errors
The theory substantially extends earlier ones by allowing dependent and/or heavy-tailed errors and the covariates, and derives exponential tail probability inequalities for dependent sub-Gaussian errors and Nagaev-type inequalities fordependent non-sub- Gaussian errors that arise from linear or non-linear processes.
Memory Properties Of Transformations Of Linear Processes And Symmetric Gini Correlation
A large class of time series processes can be modeled by linear processes, including a subset of the fractional ARIMA process. Transformation of time series is one of the most popular topics in
Exact moderate and large deviations for linear random fields
The methods of Peligrad et al. (2014) are extended and exact moderate and large deviation asymptotics for linear random fields with independent innovations are established.
Memory properties of transformations of linear processes
In this paper, we study the memory properties of transformations of linear processes. Dittmann and Granger (J Econ 110:113–133, 2002) studied the polynomial transformations of Gaussian FARIMA(0, d,
Limit theorems for linear random fields with innovations in the domain of attraction of a stable law
Cramer type moderate deviations for random fields and mutual information estimation for mixed-pair random variables
In this dissertation we first study Cramér type moderate deviation for partial sums of random fields by applying the conjugate method. In 1938 Cramér published his results on large deviations of sums
Beyond Sub-Gaussian Noises: Sharp Concentration Analysis for Stochastic Gradient Descent
The Nagaev type high-probability upper bounds for the estimation errors of averaged stochastic gradient descent (ASGD) in a linear model are obtained and it is proved that the ASGD estimate achieves an O( √ log(1/δ)/T + (δT q−1)−1/q) error rate with probability at least 1− δ, where q > 2 controls the tail of the gradient noise.
A local limit theorem for linear random fields
In this article, we establish a local limit theorem for linear fields of random variables constructed from i.i.d. innovations each with finite second moment. When the coefficients are absolutely


The supremum of a negative drift random walk with dependent heavy-tailed steps
Many important probabilistic models in queuing theory, insurance and finance deal with partial sums of a negative mean stationary process (a negative drift random walk), and the law of the supremum
We obtain asymptotic expansions for probabilities of moderate deviations for stationary causal processes. The imposed dependence conditions are easily verifiable and they are directly related to the
On linear processes with dependent innovations
Large Deviations of Heavy-Tailed Sums with Applications in Insurance
First we give a short review of large deviation results for sums of i.i.d. random variables. The main emphasis is on heavy-tailed distributions. We stress more the methodology than the detailed
We consider the unit root testing problem with errors being nonlinear transforms of linear processes+When the linear processes are long-range dependent, the asymptotic distributions in the unit root
iences, mixing sequences or associated sequences. The results are important in analyzing the asymptotical properties of some estimators as well as of linear processes. �4 1. Introduction. Let be a
Time Series with Long Memory
Long memory processes constitute a broad class of models for stationary and nonstationary time series data in economics, finance, and other fields. Their key feature is persistence, with high
On probabilities of moderate deviations of sums for independent random variables
We investigate asymptotics of probabilities of moderate deviations and their logarithms for an array of row-wise independent random variables with finite variations and finite one-sided moments of
Moderate deviations for Markovian occupation times
An Asymptotic Expansion for Probabilities of Moderate Deviations for Multivariate Martingales
We derive formulae for probabilities of large deviations in a moderate range for multivariate martingales. Although we give an elementary proof for univariate martingales, there is no elementary