arXiv Science⌕ Search

arXiv subjects

Yeor Hafouta

Publications and source records attributed to Yeor Hafouta.

At least 19 recordsLinked to original sources

Local Limit Theorems and Edgeworth Expansions for some classes of inhomogeneous Markov chains

In this note, we prove local central limit theorems and expansions in the central limit theorem for inhomogeneous Markov chains satisfying a one-step lower ellipticity condition. As for the local limit theorem, in the general inhomogeneous case, compared with \cite{DS}, we are able to prove results without two-sided ellipticity, although only in the one-step case. Compared with \cite{PeligradLLT1,PeligradLLT2}, we are able to prove local limit theorems without a lower $ψ$-mixing condition. Concerning Edgeworth expansions, compared with \cite{DHPTRF1}, we are able to get results without two-sided ellipticity (under some sequential Diophantine conditions). Our conditions are then verified for Markov chains in a random dynamical environment with a one-sided Doeblin condition under very mild assumptions. For Markov chains in a random environment, we also consider the more general case of $ϕ$-mixing chains and then, under some mild periodicity assumptions on the environment, we prove local central limit theorems and first-order Edgeworth expansions (without any ellipticity assumptions), matching the results for deterministic homogeneous chains. When the chain is driven by a topologically mixing subshift of finite type, we obtain high-order expansions for $ϕ$-mixing chains under some Diophantine conditions of a deterministic system generated by a periodic point. We then apply our results to bounded additive functionals of posterior states in disordered quantum trajectories and establish local central limit theorems and first-order Edgeworth expansions.

math.PR↗

Optimal CLT error bounds in the second order Wasserstein metric via Stein's method for weakly dependent sequences

We use Stein's method to prove general upper bounds on the second order Wasserstein distance in the CLT for general dependent partial sums $S_N=X_1+\ldots +X_N$. Then we use these estimates to provide optimal $O(N^{-1/2})$ rates for a variety of chaotic dynamical systems and $ϕ$- and $ψ$-mixing sequences. Our results for $ψ$-mixing sequences are obtained under the fourth moment condition, which matches the best known results in the iid case. This is the first time that optimal CLT rates of any kind are obtained for unbounded weakly stationary variables under general mixing conditions. We then apply this result and obtain optimal CLT rates $O\left((\text{Var}(S_N))^{-1/2}\right)$ for non-stationary $ψ$-mixing sequences without any growth rates on $\text{Var}(S_N)$. We also obtain close to optimal CLT rates for nonconventional sums and some classes of random dynamical systems with non-uniform decay of correlations

math.PR↗

Limit Theorems for Disordered Quantum Trajectories

We study discrete-time quantum trajectories of a finite-dimensional system in a disordered environment, where the instrument applied at each step is determined by an invertible measure-preserving dynamical system. For general quantum instruments on a standard Borel outcome space, we construct the quenched probability law on sequences of measurement outcomes for a fixed realization of the disorder and a given initial state. Under a quenched Doeblin condition with a uniform deterministic minorization constant and an environment-dependent minorizing probability measure, we establish the existence and uniqueness of an equivariant family of random posterior laws and identify its barycenter as the unique dynamically stationary state for the associated non-selective channel cocycle. If the environment is ergodic, we prove a pathwise ergodic theorem valid for every measurable choice of initial state: for almost every realization of the disorder, time averages of posterior states converge almost surely under the corresponding quenched law to the disorder average of the unique dynamically stationary state. For an ergodic environment, we establish quenched variance asymptotics and, when the asymptotic variance is positive, a central limit theorem and standardized Berry--Esseen bounds for additive functionals generated by bounded measurable real-valued functions of the environment and posterior state. These results hold for every measurable initial state and almost every disorder realization. We then show that the stationary annealed joint process of instruments and posterior states inherits the $α$-mixing of the instrument process, up to an exponentially decaying term. Finally, we provide classes of examples involving perfect and imperfect measurements, including models with continuous or finite outcome spaces, that satisfy the standing Doeblin condition.

math.PR↗

Berry--Esseen theorems, LLT, Edgeworth expansions, and large deviations for Locally Stationary Markov Chains

Locally stationary Markov chains have attracted a lot of attention in statistics in the past three decades. In this paper we prove a variety of limit theorems for partial sums generated by such chains. We first provide explicit formulas for the asymptotic mean and variance (and also higher moments), and obtain optimal convergence rates towards them. We then prove a Berry--Esseen theorem, a local central limit theorem, Edgeworth expansions, and large and moderate deviations principles. Our approach involves a parametric Perron--Frobenius theorem, which is proved using the theory of complex (Hilbert) projective metrics developed in \cite{Rugh,Dubois}, together with local approximation arguments and ideas in \cite{dolgopyat2023berry}.

math.PR↗

Generalized Edgeworth expansions for integer-valued additive functionals of uniformly elliptic Markov chains

We obtain asymptotic expansions for probabilities $\bbP(S_N=k)$ of partial sums of uniformly bounded integer-valued functionals $\DS S_N=\sum_{n=1}^N f_n(X_n)$ of uniformly elliptic inhomogeneous Markov chains. The expansions involve products of polynomials and trigonometric polynomials, and they hold without additional assumptions. As an application of the explicit formulas of the trigonometric polynomials, we relate existence of the standard Edgeworth expansions of order $r$ to the rate of equidistributions of $S_N$ modulo $m$ for small positive integers $m.$

math.PR↗

Thermodynamic Formalism Out of Equilibrium Part III: Local Limit Theorem and Statistical Properties

We study limit theorems for random dynamical systems. We recast the random dynamics via the skew-product of the two-point motion (the Varadhan trick). Using this approach we prove several statistical properties for the system, including quenched central limit theorems with rates, quenched local limit theorem, large deviations, and decay of correlations. Then, in applications we show how to verify the general conditions using spectral properties of appropriate operators. A key application is the study of effectively expanding on average random diffeomorphisms (in any dimension, which are allowed to be dissipative). In particular, this is the first instance of proving local limit theorems in such a broad setting, without imposing restrictions on the dynamics. We provide several new examples.

math.DS↗

Decay of correlations and limit theorems for random intermittent maps

In this paper, we revisit the problem of polynomial memory loss and the central limit theorem for time-dependent LSV maps. More precisely, we show that for random LSV maps corresponding to a random parameter beta() we obtain quenched memory loss, decay of correlations, central limit theorems with rates, moment bounds and almost sure invariance principles (ASIP) when the essential infimum of beta() is less than 1/5 and the driving process (i.e. random environment) is mixing sufficiently fast. In [59, Corollary 3.8] the ASIP was obtained for ergodic driving systems when the essential supremum of \b{eta} is less than 1/2. As will be elaborated in Section 1, restrictions on the essential infimum are more natural in our context. Our results have an abstract form which we believe could be useful in other circumstances, as will be elaborated in a future work.

math.DS↗

Local limit theorem and Edgeworth expansions for inhomogeneous random walks on $GL(d,\mathbb R)$

We prove a non-lattice local central limit theorem and Edgeworth expansions for the logarithm of the norms of products of invertible independent random matrices. Our conditions include a contraction assumption, an assumption that supports of the matrices are ``large enough" and their distributions are sufficiently regular. As a byproduct of the proof we are also able to provide a different proof to the optimal rates in the CLT proved in \cite{MatBE}. Like in \cite{MatBE} we provide several sufficient conditions for contraction.

math.PR↗

Limit theorems for inhomogeneous $ϕ$-mixing Markov chains

We prove limit theorems for inhomogeneous $ϕ$-mixing Markov chains. In the case of uniform $ϕ$-mixing, we will focus on the purely sequential (inhomogeneous) setting, while in the non-uniform case we will focus on Markov chains in random dynamical environments. A large part of the paper is devoted to verifying our main results for Markov chains satisfying a lower Doeblin condition and for Markov chains satisfying a Dobrushin-type contraction condition. We then apply our results in the random environment case to prove estimates on random mixing times as well as limit theorems for Markovian skew products. In fact, we show that one can estimate the $α$-mixing coefficients of the coordinates of the skew product, which opens the door to a variety of limit theorems. Compared with \cite{dolgopyat2023berry}, we are able to prove optimal CLT rates for inhomogeneous Markov chains solely under $ϕ$-mixing, without any ellipticity assumptions. Compared with existing results for Markov chains in random ergodic environments, under mixing conditions on the environment, we are able to treat non-uniformly $ϕ$-mixing chains, including non-uniform Doeblin/Dobrushin conditions, while all existing results either hold for uniformly mixing chains or are formulated under hard-to-verify conditions.

math.PR↗

A few notes on the asymptotic behavior of Rademacher random multiplicative functions

Let $X_p, p\in\cP$ be a sequence of independent random variables s.t. $\bbP(X_p=\pm 1)=1/2$. Let $\te_j=\prod_{p|j}X_p$ if $j$ is square free and $\te_j=0$ otherwise. Denote $S_n=\sum_{\ell=1}^n\te_\ell$. The from this point of view proving limit theorems for $S_n$ is natural problem, since $S_n$ mimics the behavior of $e^{\sqrt{\ln(β)}}$. It is a natural guiding conjecture that $S_n/\sqrt n$ obeys the central limit theorem (CLT). However, S. Chatterjee conjectured (as expressed in \cite{[25]}) that the CLT should not hold. Chatterjee's conjecture was proved by Harper \cite{[17]}, and by now it is a direct consequence of a more recent breakthrough by Harper \cite{Har20} that $\frac{S_n}{b_n}\to 0$ in $L^1$, where $b_n=(n^{1/2}(\ln(\ln(n)))^{-1/4})u_n, u_n\to\infty$. In particular $S_n/\sqrt n\to 0$. Nevertheless, the question whether there exists a sequence $a_n=o(b_n)$ such that $S_n/a_n$ converges to some limit remains a mystery. Note that the corresponding problem in the Steinhaus Setting was recently resolved by \cite{Gor1}. In this paper make an attempt to shed some light on the convergence of $S_n/a_n$. Additionally, we obtain explicit estimates on hight moments of $S_n$ without restrictions on the size of the moment compared to $n$ like in \cite[Theorem 1.2]{Har19}, which is of independent interest. This is achieved by a martingale argument together with the Burkholder inequality, and it has applications in a natural number theoretic combinatorial problem. Using martingale techniques we will also obtain exponential concentration inequalities for $S_n$ (in the large deviations regime)

math.PR↗

Effective geometric ergodicty for Markov chains in random environment

In this short note we prove ``effective" geometric ergodicity (i.e a Perron-Frobenius theorem) for Markov chains in random mixing dynamical environment satisfying a random non-uniform version of the Doeblin condition. Effectivity here means that all the random variables involved in the random exponential rates are integrable with arbitrarily large order. This compliments \cite[Theorem 2.1]{Kifer 1996}, where ``non-effective" geometric ergodicity was obtained. From a different perspective, our result is also motivated by egrodic theory, as it can be seen as an effective version of the ``spectral" gap in the top Oseledets space in the Oseledets multiplicative ergodic theorem for the random Markov operator cocycle (when it applies). We also present applications of the effective ergodicity to rates in the (quenched) almost sure invariance principle (ASIP), exponential decay of correlations for Markovian skew products and for exponential tails for random mixing times. As a byproduct of the proof of the ASIP rates we also provide easy to verify sufficient conditions for the verification of the assumptions of \cite[Theorem 2.4]{Kifer 1998}.

math.PR↗

Non-uniform Berry-Esseen theorems for weakly dependent random variables

We obtain non-uniform Berry-Esseen type estimates for several classes of weakly dependent sequences of random variables, including uniformly elliptic inhomogeneous Markov chains, random and time-varying (partially) hyperbolic or expanding dynamical systems, products of random matrices and some classes of local statistics.

math.PR↗

Quenched and annealed linear response for some partially hyperbolic skew products

We prove quenched and annealed statistical stability, linear response, and differentiability of asymptotic moments for parametric families of partially hyperbolic skew products, with random hyperbolic maps on the fibers. The main novelty is that the base maps also depend on the parameter, which leads to different formulas in the linear response and the derivative of the asymptotic moments with respect to the parameter. Our annealed results apply to partially hyperbolic maps that are not covered in \cite{BashCastro26,Dol,DS}.

math.DS↗

Berry-Esseen Bounds for Statistics of Non-Stationary, $ϕ$-Mixing Random Variables

Using a modification of Stein's method, we generalize the results of Bentkus, G{ö}tze, and Tikhomirov \cite{bentkus1997berry} to obtain Berry-Esseen bounds for a broad class of statistics of sequences of $ϕ$-mixing, non-stationary random variables with polynomial mixing rates. %and linear variance. We then consider applications of this theorem to ensure Berry-Esseen rates for various classes of non-stationary $ϕ$-mixing random variables, including rates for a general class of processes of $ϕ$-mixing random variables satisfying an aggregate third moment bound.

math.PR↗

Effective quenched linear response for random dynamical systems

We prove ``effective'' linear response for certain classes of non-uniformly expanding random dynamical systems which are not necessarily composed in an i.i.d manner. In applications, the results are obtained for base maps with a sufficient amount of mixing. The fact that the rates are effective is then applied to obtain the differentiability of the variance in the CLT as a function of the parameter, as well as the annealed linear response. These two applications are beyond the reach of the linear response obtained in the general case, when all the random variables appearing in the bounds are only tempered. We also provide several wide examples of one-dimensional maps satisfying our conditions, as well as some higher-dimensional examples.

math.DS↗

Limit theorems for inhomogeneous random walks on $GL(d,\mathbb R)$

We prove Berry-Esseen theorems, almost sure invariance principle rates and large deviations for products of independent but not identically distributed invertible matrices with some average (logarithmic) projective contraction and uniform boundedness assumptions. We also characterize the divergence of the variance of the logarithm of the norm of the product. Our approach is based on verifying the conditions of \cite{NewBE} after reversing time.

math.PR↗

Statistical properties of Markov shifts: part II-LLT

We prove Local Central Limit Theorems (LLT) for partial sums of the form $S_n=\sum_{j=0}^{n-1}f_j(...,X_{j-1},X_j,X_{j+1},...)$, where $(X_j)$ is a Markov chains with equicontinuous conditional probabilities satisfying contraction conditions close in spirit to Dobrushin's, and some ``physicality" assumptions and $f_j$ are equicontinuous functions. Our conditions will always be in force when the chain takes values on a metric space and have uniformly bounded away from $0$ backward transition densities with respect to a measure which assigns uniform positive mass to certain ``balls". This paper complements \cite{MarShif1} where Berry-Esseen theorems, were proven for (not necessarily continuous) functions satisfying certain approximation conditions. Our results address a question posed by D. Dolgopyat and O. Sarig in \cite[Section 1.5]{DS}.

math.PR↗

Livšic regularity for random and sequential dynamics through transfer operators

We prove Livšic-type regularity results of coboundary representations for non-autonomous dynamical systems. Our results have an abstract nature and apply to several important specific situations, such as (higher-dimensional) random or sequential piecewise expanding maps and subshifts of finite type, which have applications to Markov interval maps and to finite state inhomogeneous elliptic Markov shifts, via symbolic representations. We also obtain results for some classes of non-autonomous hyperbolic systems. Our results can be seen as non-autonomous versions of a recent result obtained by Morris. However, we emphasize that our proof differs from the one mentioned previously even in the deterministic case. Finally, we show that our results provide a more relaxed characterization for having variance growth of Birkhoff sums on random and sequential dynamical systems; we show that such growth can fail only when the underlying functions are a coboundary without special restrictions on the regularity of the coboundary. For random systems, we show that this is equivalent to having a coboundary with bounded ``variation", but for sequential systems it turns out that this is no longer true, as demonstrated by examples.

math.DS↗