Showing 1–14 of 14 results for author: Mariucci, E

Nonparametric density estimation for the small jumps of Lévy processes

Authors: Céline Duval, Taher Jalal, Ester Mariucci

Abstract: We consider the problem of estimating the density of the process associated with the small jumps of a pure jump Lévy process, possibly of infinite variation, from discrete observations of one trajectory. The interest of such a question lies on the observation that even when the Lévy measure is known, the density of the increments of the small jumps of the process cannot be computed. We discuss res… ▽ More We consider the problem of estimating the density of the process associated with the small jumps of a pure jump Lévy process, possibly of infinite variation, from discrete observations of one trajectory. The interest of such a question lies on the observation that even when the Lévy measure is known, the density of the increments of the small jumps of the process cannot be computed. We discuss results both from low and high frequency observations. In a low frequency setting, assuming the Lévy density associated with the jumps larger than $\varepsilon\in (0,1]$ in absolute value is known, a spectral estimator relying on the convolution structure of the problem achieves minimax parametric rates of convergence with respect to the integrated $L_2$ loss, up to a logarithmic factor. In a high frequency setting, we remove the assumption on the knowledge of the Lévy measure of the large jumps and show that the rate of convergence depends both on the sampling scheme and on the behaviour of the Lévy measure in a neighborhood of zero. We show that the rate we find is minimax up to a log-factor. An adaptive penalized procedure is studied to select the cutoff parameter. These results are extended to encompass the case where a Brownian component is present in the Lévy process. Furthermore, we illustrate the performances of our procedures through an extensive simulation study. △ Less

Submitted 15 April, 2024; originally announced April 2024.

Comments: 31 pages, 3 figures, 4 tables

Non-asymptotic control of the cumulative distribution function of Lévy processes

Authors: Céline Duval, Ester Mariucci

Abstract: We propose non-asymptotic controls of the cumulative distribution function $P(|X_{t}|\ge \varepsilon)$, for any $t>0$, $\varepsilon>0$ and any Lévy process $X$ such that its Lévy density is bounded from above by the density of an $α$-stable type Lévy process in a neighborhood of the origin. The results presented are non-asymptotic and optimal, they apply to a large class of Lévy processes. We propose non-asymptotic controls of the cumulative distribution function $P(|X_{t}|\ge \varepsilon)$, for any $t>0$, $\varepsilon>0$ and any Lévy process $X$ such that its Lévy density is bounded from above by the density of an $α$-stable type Lévy process in a neighborhood of the origin. The results presented are non-asymptotic and optimal, they apply to a large class of Lévy processes. △ Less

Submitted 20 March, 2020; originally announced March 2020.

Decompounding discrete distributions: A non-parametric Bayesian approach

Authors: Shota Gugushvili, Ester Mariucci, Frank van der Meulen

Abstract: Suppose that a compound Poisson process is observed discretely in time and assume that its jump distribution is supported on the set of natural numbers. In this paper we propose a non-parametric Bayesian approach to estimate the intensity of the underlying Poisson process and the distribution of the jumps. We provide a MCMC scheme for obtaining samples from the posterior. We apply our method on bo… ▽ More Suppose that a compound Poisson process is observed discretely in time and assume that its jump distribution is supported on the set of natural numbers. In this paper we propose a non-parametric Bayesian approach to estimate the intensity of the underlying Poisson process and the distribution of the jumps. We provide a MCMC scheme for obtaining samples from the posterior. We apply our method on both simulated and real data examples, and compare its performance with the frequentist plug-in estimator proposed by Buchmann and Grübel. On a theoretical side, we study the posterior from the frequentist point of view and prove that as the sample size $n\rightarrow\infty$, it contracts around the `true', data-generating parameters at rate $1/\sqrt{n}$, up to a $\log n$ factor. △ Less

Submitted 20 May, 2020; v1 submitted 26 March, 2019; originally announced March 2019.

Comments: 27 pages, 7 figures

MSC Class: 62G20 (Primary); 62M30 (Secondary)

Journal ref: Scand J. Statist., 47(2):464-492, 2020

Total variation distance for discretely observed Lévy processes: a Gaussian approximation of the small jumps

Authors: Alexandra Carpentier, Céline Duval, Ester Mariucci

Abstract: It is common practice to treat small jumps of Lévy processes as Wiener noise and thus to approximate its marginals by a Gaussian distribution. However, results that allow to quantify the goodness of this approximation according to a given metric are rare. In this paper, we clarify what happens when the chosen metric is the total variation distance. Such a choice is motivated by its statistical int… ▽ More It is common practice to treat small jumps of Lévy processes as Wiener noise and thus to approximate its marginals by a Gaussian distribution. However, results that allow to quantify the goodness of this approximation according to a given metric are rare. In this paper, we clarify what happens when the chosen metric is the total variation distance. Such a choice is motivated by its statistical interpretation. If the total variation distance between two statistical models converges to zero, then no tests can be constructed to distinguish the two models which are therefore equivalent, statistically speaking. We elaborate a fine analysis of a Gaussian approximation for the small jumps of Lévy processes with infinite Lévy measure in total variation distance. Non asymptotic bounds for the total variation distance between $n$ discrete observations of small jumps of a Lévy process and the corresponding Gaussian distribution is presented and extensively discussed. As a byproduct, new upper bounds for the total variation distance between discrete observations of Lévy processes are provided. The theory is illustrated by concrete examples. △ Less

Submitted 2 April, 2019; v1 submitted 6 October, 2018; originally announced October 2018.

Comments: Important and necessary changes have been made in this new version, this version supersedes version 1

MSC Class: 60G51; 62M99 (Primary); 60E99 (Secondary)

Wasserstein and total variation distance between marginals of Lévy processes

Authors: Ester Mariucci, Markus Reiß

Abstract: We present upper bounds for the Wasserstein distance of order $p$ between the marginals of Lévy processes, including Gaussian approximations for jumps of infinite activity. Using the convolution structure, we further derive upper bounds for the total variation distance between the marginals of Lévy processes. Connections to other metrics like Zolotarev and Toscani-Fourier distances are established… ▽ More We present upper bounds for the Wasserstein distance of order $p$ between the marginals of Lévy processes, including Gaussian approximations for jumps of infinite activity. Using the convolution structure, we further derive upper bounds for the total variation distance between the marginals of Lévy processes. Connections to other metrics like Zolotarev and Toscani-Fourier distances are established. The theory is illustrated by concrete examples and an application to statistical lower bounds. △ Less

Submitted 16 July, 2018; v1 submitted 7 October, 2017; originally announced October 2017.

Comments: 32 pages. To appear in Electronic Journal of Statistics

A Bayesian nonparametric approach to log-concave density estimation

Authors: Ester Mariucci, Kolyan Ray, Botond Szabo

Abstract: The estimation of a log-concave density on $\mathbb{R}$ is a canonical problem in the area of shape-constrained nonparametric inference. We present a Bayesian nonparametric approach to this problem based on an exponentiated Dirichlet process mixture prior and show that the posterior distribution converges to the log-concave truth at the (near-) minimax rate in Hellinger distance. Our proof proceed… ▽ More The estimation of a log-concave density on $\mathbb{R}$ is a canonical problem in the area of shape-constrained nonparametric inference. We present a Bayesian nonparametric approach to this problem based on an exponentiated Dirichlet process mixture prior and show that the posterior distribution converges to the log-concave truth at the (near-) minimax rate in Hellinger distance. Our proof proceeds by establishing a general contraction result based on the log-concave maximum likelihood estimator that prevents the need for further metric entropy calculations. We also present two computationally more feasible approximations and a more practical empirical Bayes approach, which are illustrated numerically via simulations. △ Less

Submitted 30 November, 2018; v1 submitted 28 March, 2017; originally announced March 2017.

Comments: 39 pages, 17 figures. Simulation studies were significantly expanded and one more theorem has been added

MSC Class: 62G07; 62G20

Journal ref: Bernoulli 26 (2020), no. 2, 1070-1097

Spectral-free estimation of Lévy densities in high-frequency regime

Authors: Céline Duval, Ester Mariucci

Abstract: We construct an estimator of the Lévy density of a pure jump Lévy process, possibly of infinite variation, from the discrete observation of one trajectory at high frequency. The novelty of our procedure is that we directly estimate the Lévy density relying on a pathwise strategy, whereas existing procedures rely on spectral techniques. By taking advantage of a compound Poisson approximation, we ci… ▽ More We construct an estimator of the Lévy density of a pure jump Lévy process, possibly of infinite variation, from the discrete observation of one trajectory at high frequency. The novelty of our procedure is that we directly estimate the Lévy density relying on a pathwise strategy, whereas existing procedures rely on spectral techniques. By taking advantage of a compound Poisson approximation, we circumvent the use of spectral techniques and in particular of the Lévy--Khintchine formula. A linear wavelet estimator is built and its performance is studied in terms of $L_p$ loss functions, $p\geq 1$, over Besov balls. We recover classical nonparametric rates for finite variation Lévy processes and for a large nonparametric class of symmetric infinite variation Lévy processes. We show that the procedure is robust when the estimation set gets close to the critical value 0 and also discuss its robustness to the presence of a Brownian part. △ Less

Submitted 3 April, 2020; v1 submitted 28 February, 2017; originally announced February 2017.

Comments: 33 pages. Several improvements made on the generality of the results

Le cam theory on the comparison of statistical models

Authors: Ester Mariucci

Abstract: We recall the main concepts of the Le Cam theory of statistical experiments , especially the notion of Le Cam distance and its properties. We also review classical tools for bounding such a distance before presenting some examples. A proof of the classical equivalence result between density estimation problems and Gaussian white noise models will be analyzed. We recall the main concepts of the Le Cam theory of statistical experiments , especially the notion of Le Cam distance and its properties. We also review classical tools for bounding such a distance before presenting some examples. A proof of the classical equivalence result between density estimation problems and Gaussian white noise models will be analyzed. △ Less

Submitted 11 May, 2016; originally announced May 2016.

Asymptotic equivalence for density estimation and gaussian white noise: An extension

Authors: Ester Mariucci

Abstract: The aim of this paper is to present an extension of the well-known as-ymptotic equivalence between density estimation experiments and a Gaussian white noise model. Our extension consists in enlarging the nonparametric class of the admissible densities. More precisely, we propose a way to allow densities defined on any subinterval of R, and also some discontinuous or unbounded densities are conside… ▽ More The aim of this paper is to present an extension of the well-known as-ymptotic equivalence between density estimation experiments and a Gaussian white noise model. Our extension consists in enlarging the nonparametric class of the admissible densities. More precisely, we propose a way to allow densities defined on any subinterval of R, and also some discontinuous or unbounded densities are considered (so long as the discontinuity and unboundedness patterns are somehow known a priori). The concept of equivalence that we shall adopt is in the sense of the Le Cam distance between statistical models. The results are constructive: all the asymptotic equivalences are established by constructing explicit Markov kernels. △ Less

Submitted 17 March, 2015; originally announced March 2015.

Comments: 11 pages. arXiv admin note: text overlap with arXiv:1503.04530

Asymptotic equivalence for pure jump Lévy processes with unknown Lévy density and Gaussian white noise

Authors: Ester Mariucci

Abstract: The aim of this paper is to establish a global asymptotic equivalence between the experiments generated by the discrete (high frequency) or continuous observation of a path of a L{é}vy process and a Gaussian white noise experiment observed up to a time T, with T tending to $\infty$. These approximations are given in the sense of the Le Cam distance, under some smoothness conditions on the unknown… ▽ More The aim of this paper is to establish a global asymptotic equivalence between the experiments generated by the discrete (high frequency) or continuous observation of a path of a L{é}vy process and a Gaussian white noise experiment observed up to a time T, with T tending to $\infty$. These approximations are given in the sense of the Le Cam distance, under some smoothness conditions on the unknown L{é}vy density. All the asymptotic equivalences are established by constructing explicit Markov kernels that can be used to reproduce one experiment from the other. △ Less

Submitted 12 August, 2015; v1 submitted 16 March, 2015; originally announced March 2015.

Comments: 50 pages. The definition of the parameter space has changed and some proofs have been expanded and corrected

Asymptotic equivalence of discretely observed diffusion processes and their Euler scheme: small variance case

Authors: Ester Mariucci

Abstract: This paper establishes the global asymptotic equivalence, in the sense of the Le Cam $Δ$-distance, between scalar diffusion models with unknown drift function and small variance on the one side, and nonparametric autoregressive models on the other side. The time horizon $T$ is kept fixed and both the cases of discrete and continuous observation of the path are treated. We allow non constant diffus… ▽ More This paper establishes the global asymptotic equivalence, in the sense of the Le Cam $Δ$-distance, between scalar diffusion models with unknown drift function and small variance on the one side, and nonparametric autoregressive models on the other side. The time horizon $T$ is kept fixed and both the cases of discrete and continuous observation of the path are treated. We allow non constant diffusion coefficient, bounded but possibly tending to zero. The asymptotic equivalences are established by constructing explicit equivalence map**s. △ Less

Submitted 4 March, 2015; v1 submitted 13 November, 2014; originally announced November 2014.

Comments: 21 pages

Asymptotic equivalence for inhomogeneous jump diffusion processes and white noise

Authors: Ester Mariucci

Abstract: We prove the global asymptotic equivalence between the experiments generated by the discrete (high frequency) or continuous observation of a path of a time inhomogeneous jump-diffusion process and a Gaussian white noise experiment. Here, the considered parameter is the drift function, and we suppose that the observation time $T$ tends to $\infty$. The approximation is given in the sense of the Le… ▽ More We prove the global asymptotic equivalence between the experiments generated by the discrete (high frequency) or continuous observation of a path of a time inhomogeneous jump-diffusion process and a Gaussian white noise experiment. Here, the considered parameter is the drift function, and we suppose that the observation time $T$ tends to $\infty$. The approximation is given in the sense of the Le Cam $Δ$-distance, under smoothness conditions on the unknown drift function. These asymptotic equivalences are established by constructing explicit Markov kernels that can be used to reproduce one experiment from the other. △ Less

Submitted 22 March, 2015; v1 submitted 2 May, 2014; originally announced May 2014.

Comments: 20 pages; to appear on ESAIM: P\&S. In this version there are some improvements in the exposition following the reports suggestions

L_1-distance for additive processes with time-homogeneous Lévy measures

Authors: Pierre Etore, Ester Mariucci

Abstract: We give an explicit bound for the $L_1$-distance between two additive processes of local characteristics $(f_j(\cdot),σ^2(\cdot),ν_j)$, $j = 1,2$. The cases $σ=0$ and $σ> 0$ are both treated. We allow $ν_1$ and $ν_2$ to be equivalent time-homogeneous Lévy measures, possibly with infinite variation. Some examples of possible applications are discussed. We give an explicit bound for the $L_1$-distance between two additive processes of local characteristics $(f_j(\cdot),σ^2(\cdot),ν_j)$, $j = 1,2$. The cases $σ=0$ and $σ> 0$ are both treated. We allow $ν_1$ and $ν_2$ to be equivalent time-homogeneous Lévy measures, possibly with infinite variation. Some examples of possible applications are discussed. △ Less

Submitted 15 May, 2014; v1 submitted 30 April, 2014; originally announced April 2014.

Comments: 9 pages; extended introduction and added references

Asymptotic equivalence of jumps Lévy processes and their discrete counterpart

Authors: Pierre Étoré, Sana Louhichi, Ester Mariucci

Abstract: We establish the global asymptotic equivalence between a pure jumps Lévy process $\{X_t\}$ on the time interval $[0,T]$ with unknown Lévy measure $ν$ belonging to a non-parametric class and the observation of $2m^2$ Poisson independent random variables with parameters linked with the Lévy measure $ν$. The equivalence result is asymptotic as $m$ tends to infinity. The time $T$ is kept fixed and the… ▽ More We establish the global asymptotic equivalence between a pure jumps Lévy process $\{X_t\}$ on the time interval $[0,T]$ with unknown Lévy measure $ν$ belonging to a non-parametric class and the observation of $2m^2$ Poisson independent random variables with parameters linked with the Lévy measure $ν$. The equivalence result is asymptotic as $m$ tends to infinity. The time $T$ is kept fixed and the sample path is continuously observed. This result justifies the idea that, from a statistical point of view, knowing how many jumps fall into a grid of intervals gives asymptotically the same amount of information as observing $\{X_t\}$. △ Less

Submitted 19 September, 2013; v1 submitted 29 May, 2013; originally announced May 2013.

Comments: Shorter version focusing on the statistical analysis of the Lévy measure. A new example has been added