arXiv subjects
Publications and source records attributed to Emanuele Dolera.
In the setting of dominated statistical models, we provide conditions yielding strong continuity of the posterior distribution with respect to the observed data. We show some applications, with special focus on exponential models.
Consider $n$ independent measurements, with the additional information of the times at which measurements are performed. This paper deals with testing statistical hypotheses when $n$ is large and only a small amount of observations concentrated in short time intervals are relevant to the study. We define a testing procedure in terms of multiple likelihood ratio (LR) statistics obtained by splitting the observations into groups, and in accordance with the following principles: P1) each LR statistic is formed by gathering the data included in $G$ consecutive vectors of observations, where $G$ is a suitable time window defined a priori with respect to an arbitrary choice of the `origin of time'; P2) the null statistical hypothesis is rejected only if at least $k$ LR statistics are sufficiently small, for a suitable choice of $k$. We show that the application of the classical Wilks' theorem may be affected by the arbitrary choice of the "origin of time", in connection with P1). We then introduce a Wilks' theorem for grouped data which leads to a testing procedure that overcomes the problem of the arbitrary choice of the `origin of time', while fulfilling P1) and P2). Such a procedure is more powerful than the corresponding procedure based on Wilks' theorem.
This paper deals with the problem of quantifying the approximation a probability measure by means of an empirical (in a wide sense) random probability measure, depending on the first n terms of a sequence of random elements. In Section 2, one studies the range of oscillation near zero of the Wasserstein distance $\ud^{(p)}_{\pms}$ between $\pfrak_0$ and $\hat{\pfrak}_n$, assuming that the $\xitil_i$'s are i.i.d. with $\pfrak_0$ as common law. Theorem 2.3 deals with the case in which $\pfrak_0$ is fixed as a generic element of the space of all probability measures on $(\rd, \mathscr{B}(\rd))$ and $\hat{\pfrak}_n$ coincides with the empirical measure. In Theorem 2.4 (Theorem 2.5, respectively) \pfrak_0 is a d-dimensional Gaussian distribution (an element of a distinguished type of statistical exponential family, respectively) and $\hat{\pfrak}_n$ is another $d$-dimensional Gaussian distribution with estimated mean and covariance matrix (another element of the same family with an estimated parameter, respectively). These new results improve on allied recent works (see, e.g., [31]) since they also provide uniform bounds with respect to $n$, meaning that the finiteness of the p-moment of the random variable $\sup_{n \geq 1} b_n \ud^{(p)}_{\pms}(\pfrak_0, \hat{\pfrak}_n)$ is proved for some suitable diverging sequence b_n of positive numbers. In Section 3, under the hypothesis that the $\xitil_i$'s are exchangeable, one studies the range of the random oscillation near zero of the Wasserstein distance between the conditional distribution--also called posterior--of the directing measure of the sequence, given $\xitil_1, \dots, \xitil_n$, and the point mass at $\hat{\pfrak}_n$. In a similar vein, a bound for the approximation of predictive distributions is given. Finally, Theorems from 3.3 to 3.5 reconsider Theorems from 2.3 to 2.5, respectively, according to a Bayesian perspective.
Given a sequence ξ_1, ξ_2,... of X-valued, exchangeable random elements, let q(ξ^(n)) and p_m(ξ^(n)) stand for posterior and predictive distribution, respectively, given ξ^(n) = (ξ_1,..., ξ_n). We provide an upper bound for limsup b_n d_[[X]](q(ξ^(n)), δ_\empiricn) and limsup b_n d_[X^m](p_m(ξ^(n)), \empiricn^m), where \empiricn is the empirical measure, b_n is a suitable sequence of positive numbers increasing to +\infty, d_[[X]] and d_[X^m] denote distinguished weak probability distances on [[X]] and [X^m], respectively, with the proviso that [S] denotes the space of all probability measures on S. A characteristic feature of our work is that the aforesaid bounds are established under the law of the ξ_n's, unlike the more common literature on Bayesian consistency, where they are studied with respect to product measures (p_0)^\infty, as p_0 varies among the admissible determinations of a random probability measure.
This paper provides a generalization of a classical result obtained by Wilks about the asymptotic behavior of the likelihood ratio. The new results deal with the asymptotic behavior of the joint distribution of a vector of likelihood ratios which turn out to be stochastically dependent.
This paper proves the existence of weak solutions to the spatially homogeneous Boltzmann equation for Maxwellian molecules, when the initial data are chosen from the space of all Borel probability measures on R^3 with finite second moments and the (angular) collision kernel satisfies a very weak cutoff condition. Conservation of momentum and energy is also proved for these weak solutions, without resorting to any boundedness of the entropy.
The present work provides a definitive answer to the problem of quantifying relaxation to equilibrium of the solution to the spatially homogeneous Boltzmann equation for Maxwellian molecules. The beginning of the story dates back to a pioneering work by Hilbert, who first formalized the concept of linearization of the collision operator and pointed out the importance of its eigenvalues with respect to a certain asymptotic behavior of the Boltzmann equation. Under really mild conditions on initial data - close to being necessary - and a weak, physically consistent, angular cutoff hypothesis, our main result (Theorem 1.1) contains the first precise statement that the total variation distance between the solution and the limiting Maxwellian distribution admits an upper bound of the form C e^{Lambda_b t}, Lambda_b being the least negative of the aforesaid eigenvalues and C a constant which depends only on a few simple numerical characteristics (e.g. moments) of the initial datum. The validity of this quantification was conjectured, about fifty years ago, in a paper by Henry P. McKean but, in spite of several attempts, the best answer known up to now consists in a bound with a rate which can be made arbitrarily close to Lambda_b, to the cost of the "explosion" of the constant C. Moreover, its deduction is subject to restrictive hypotheses on the initial datum, besides the Grad angular cutoff condition. As to the proof of our results, we have taken as point of reference an analogy between the problem of convergence to equilibrium and the central limit theorem of probability theory, highlighted by McKean. Our work represents in fact a confirmation of this analogy, since the techniques we develop here crucially rely on certain formulations of the central limit theorem.
This note deals with the long-time behavior of the solution to the spatially homogeneous Boltzmann equation for Maxwellian molecules, when the initial datum belongs to a suitable neighborhood of the Maxwellian equilibrium. In particulary, it contains a quantification of the rate of exponential convergence, obtained by simple arguments.
Consider the homogeneous Boltzmann equation for Maxwellian molecules. We provide a new representation for its solution in the form of expectation of a random probability measure M. We also prove that the Fourier transform of M is a conditional characteristic function of a sum of independent random variables, given a suitable sigma-algebra. These facts are then used to prove a CLT for Maxwellian molecules, that is the statement of a necessary and sufficient condition for the weak convergence of the solution of the equation. Such a condition reduces to the finiteness of the second moment of the initial distribution μ_0. As a further application, we give a refinement of some inequalities, due to Elmroth, concerning the evolution of the moments of the solution.
In Dolera, Gabetta and Regazzini [Ann. Appl. Probab. 19 (2009) 186-201] it is proved that the total variation distance between the solution $f(\cdot,t)$ of Kac's equation and the Gaussian density $(0,σ^2)$ has an upper bound which goes to zero with an exponential rate equal to -1/4 as $t\to+\infty$. In the present paper, we determine a lower bound which decreases exponentially to zero with this same rate, provided that a suitable symmetrized form of $f_0$ has nonzero fourth cumulant $κ_4$. Moreover, we show that upper bounds like $\bar{C}_δe^{-({1/4})t}ρ_δ(t)$ are valid for some $ρ_δ$ vanishing at infinity when $\int_{\mathbb{R}}|v|^{4+δ}f_0(v)\,dv<+\infty$ for some $δ$ in $[0,2[$ and $κ_4=0$. Generalizations of this statement are presented, together with some remarks about non-Gaussian initial conditions which yield the insuperable barrier of -1 for the rate of convergence.
Let $f(\cdot,t)$ be the probability density function which represents the solution of Kac's equation at time $t$, with initial data $f_0$, and let $g_σ$ be the Gaussian density with zero mean and variance $σ^2$, $σ^2$ being the value of the second moment of $f_0$. This is the first study which proves that the total variation distance between $f(\cdot,t)$ and $g_σ$ goes to zero, as $t\to +\infty$, with an exponential rate equal to -1/4. In the present paper, this fact is proved on the sole assumption that $f_0$ has finite fourth moment and its Fourier transform $φ_0$ satisfies $|φ_0(ξ)|=o(|ξ|^{-p})$ as $|ξ|\to+\infty$, for some $p>0$. These hypotheses are definitely weaker than those considered so far in the state-of-the-art literature, which in any case, obtains less precise rates.