DeepAI AI Chat
Log In Sign Up

A Finite-Sample Deviation Bound for Stable Autoregressive Processes

by   Rodrigo A. González, et al.

In this paper, we study non-asymptotic deviation bounds of the least squares estimator in Gaussian AR(n) processes. By relying on martingale concentration inequalities and a tail-bound for χ^2 distributed variables, we provide a concentration bound for the sample covariance matrix of the process output. With this, we present a problem-dependent finite-time bound on the deviation probability of any fixed linear combination of the estimated parameters of the AR(n) process. We discuss extensions and limitations of our approach.


page 1

page 2

page 3

page 4


Finite sample deviation and variance bounds for first order autoregressive processes

In this paper, we study finite-sample properties of the least squares es...

Least absolute deviation estimation for AR(1) processes with roots close to unity

We establish the asymptotic theory of least absolute deviation estimator...

Risk Bounds for Infinitely Divisible Distribution

In this paper, we study the risk bounds for samples independently drawn ...

Bump detection in the presence of dependency: Does it ease or does it load?

We provide the asymptotic minimax detection boundary for a bump, i.e. an...

Finite impulse response models: A non-asymptotic analysis of the least squares estimator

We consider a finite impulse response system with centered independent s...

Instance-dependent uniform tail bounds for empirical processes

We formulate a uniform tail bound for empirical processes indexed by a c...

Finite-time Identification of Stable Linear Systems: Optimality of the Least-Squares Estimator

We provide a new finite-time analysis of the estimation error of stable ...