Local linear estimation for stochastic processes driven by \alpha -stable L\acute{\mathbf{e}}vy motion

The $\alpha $ α -stable L $\acute{\mathrm{e}}$ e ´ vy motion together with the Poisson process and Brownian motion are the most important examples of L $\acute{\mathrm{e}}$ e ´ vy processes, which form the first class of stochastic processes being studied in the modern spirit. In this paper, the stochastic processes driven by $\alpha $ α -stable L $\acute{\mathrm{e}}$ e ´ vy motion are considered, local linear estimator of the drift function for these processes is discussed. Under mild conditions, we derive consistency of the local linear estimator of the drift function. The performance of the proposed estimator is assessed by simulation study.     

Distributions of the maximum likelihood and minimum contrast estimators associated with the fractional Ornstein–Uhlenbeck process

We discuss some inference problems associated with the fractional Ornstein–Uhlenbeck (fO–U) process driven by the fractional Brownian motion (fBm). In particular, we are concerned with the estimation of the drift parameter, assuming that the Hurst parameter $H$ is known and is in $[1/2, 1)$ . Under this setting we compute the distributions of the maximum likelihood estimator (MLE) and the minimum contrast estimator (MCE) for the drift parameter, and explore their distributional properties by paying attention to the influence of $H$ and the sampling span $M$ . We also deal with the ordinary least squares estimator (OLSE) and examine the asymptotic relative efficiency. It is shown that the MCE is asymptotically efficient, while the OLSE is inefficient. We also consider the unit root testing problem in the fO–U process and compute the power of the tests based on the MLE and MCE.     

Empirical likelihood bivariate nonparametric maximum likelihood estimator with right censored data

This article considers the estimation for bivariate distribution function (d.f.) \(F_0(t, z)\) of survival time \(T\) and covariate variable \(Z\) based on bivariate data where \(T\) is subject to right censoring. We derive the empirical likelihood-based bivariate nonparametric maximum likelihood estimator \(\hat{F}_n(t,z)\) for \(F_0(t,z)\) , which has an explicit expression and is unique in the sense of empirical likelihood. Other nice features of \(\hat{F}_n(t,z)\) include that it has only nonnegative probability masses, thus it is monotone in bivariate sense. We show that under \(\hat{F}_n(t,z)\) , the conditional d.f. of \(T\) given \(Z\) is of the same form as the Kaplan–Meier estimator for the univariate case, and that the marginal d.f. \(\hat{F}_n(\infty ,z)\) coincides with the empirical d.f. of the covariate sample. We also show that when there is no censoring, \(\hat{F}_n(t,z)\) coincides with the bivariate empirical d.f. For discrete covariate \(Z\) , the strong consistency and weak convergence of \(\hat{F}_n(t,z)\) are established. Some simulation results are presented.     

On Estimating the Asymptotic Variance of Stationary Point Processes

We investigate a class of kernel estimators $\widehat{\sigma}^2_n$ of the asymptotic variance σ ² of a d-dimensional stationary point process $\Psi = \sum_{i\ge 1}\delta_{X_i}$ which can be observed in a cubic sampling window $W_n = [-n,n]^d\,$ . σ ² is defined by the asymptotic relation $Var(\Psi(W_n)) \sim \sigma^2 \,(2n)^d$ (as n →? ∞) and its existence is guaranteed whenever the corresponding reduced covariance measure $\gamma^{(2)}_{red}(\cdot)$ has finite total variation. Depending on the rate of decay (polynomially or exponentially) of the total variation of $\gamma^{(2)}_{red}(\cdot)$ outside of an expanding ball centered at the origin, we determine optimal bandwidths b _n (up to a constant) minimizing the mean squared error of $\widehat{\sigma}^2_n$ . The case when $\gamma^{(2)}_{red}(\cdot)$ has bounded support is of particular interest. Further we suggest an isotropised estimator $\widetilde{\sigma}^2_n$ suitable for motion-invariant point processes and compare its properties with $\widehat{\sigma}^2_n$ . Our theoretical results are illustrated and supported by a simulation study which compares the (relative) mean squared errors of $\widehat{\sigma}^2_n$ for planar Poisson, Poisson cluster, and hard-core point processes and for various values of n b _n .     

Donsker-type theorems for nonparametric maximum likelihood estimators

Let ${\mathcal{P}}$ be a nonparametric probability model consisting of smooth probability densities and let ${\hat{p}_{n}}$ be the corresponding maximum likelihood estimator based on n independent observations each distributed according to the law ${\mathbb{P}}$ . With $\hat{\mathbb{P}}_{n}$ denoting the measure induced by the density ${\hat{p}_{n}}$ , define the stochastic process ${\hat{\nu}}_{n}: f\longmapsto \sqrt{n} \int fd({\hat{\mathbb{P}}}_{n} -\mathbb{P})$ where f ranges over some function class ${\mathcal{F}}$ . We give a general condition for Donsker classes ${\mathcal{F}}$ implying that the stochastic process $\hat{\nu}_{n}$ is asymptotically equivalent to the empirical process in the space ${\ell ^{\infty }(\mathcal{F})}$ of bounded functions on ${ \mathcal{F}}$ . This implies in particular that $\hat{\nu}_{n}$ converges in law in ${\ell ^{\infty }(\mathcal{F})}$ to a mean zero Gaussian process. We verify the general condition for a large family of Donsker classes ${\mathcal{ F}}$ . We give a number of applications: convergence of the probability measure ${\hat{\mathbb{P}}_{n}}$ to ${\mathbb{P}}$ at rate ${\sqrt{n}}$ in certain metrics metrizing the topology of weak(-star) convergence; a unified treatment of convergence rates of the MLE in a continuous scale of Sobolev-norms; ${\sqrt{n}}$ -efficient estimation of nonlinear functionals defined on ${\mathcal{P}}$ ; limit theorems at rate ${\sqrt{n}}$ for the maximum likelihood estimator of the convolution product ${\mathbb{P\ast P}}$ .     

Bandwidth selection for the estimation of transition probabilities in the location-scale progressive three-state model

Times between consecutive events are often of interest in medical studies. Usually the events represent different states of the disease process and are modeled using multi-state models. This paper introduces and studies a feasible estimation method for the transition probabilities in a progressive three-state model. We assume that the vector of gap times $(T_1,T_2)$ satisfies a nonparametric location-scale regression model $T_2=m(T_1)+\sigma (T_1)\epsilon $ , where the functions $m$ and $\sigma $ are ‘smooth’, and $\epsilon $ is independent of $T_1$ . Under this model, Van Keilegom et al. (J Stat Plan Inference 141:1118–1131, 2011) proposed estimators of the transition probabilities. However, the important issue of automatic bandwidth choice in this setting has not been examined, making the analysis of real datasets rather difficult. In this paper, we study the performance of their estimator in practice, we propose some modifications and study practical issues related to the implementation of the estimator, which involves the choice of an appropriate bandwidth. In an extensive simulation study the good performance of the method is shown. Simulations also demonstrate that the proposed estimator compares favorably with alternative estimators. Furthermore, the proposed methodology is illustrated with a real database on breast cancer.     

Estimating the intensity of a random measure by histogram type estimators

The purpose of this paper is to estimate the intensity of some random measure N on a set ${\mathcal{X}}$ by a piecewise constant function on a finite partition of ${\mathcal{X}}$ . Given a (possibly large) family ${\mathcal{M}}$ of candidate partitions, we build a piecewise constant estimator (histogram) on each of them and then use the data to select one estimator in the family. Choosing the square of a Hellinger-type distance as our loss function, we show that each estimator built on a given partition satisfies an analogue of the classical squared bias plus variance risk bound. Moreover, the selection procedure leads to a final estimator satisfying some oracle-type inequality, with, as usual, a possible loss corresponding to the complexity of the family ${\mathcal{M}}$ . When this complexity is not too high, the selected estimator has a risk bounded, up to a universal constant, by the smallest risk bound obtained for the estimators in the family. For suitable choices of the family of partitions, we deduce uniform risk bounds over various classes of intensities. Our approach applies to the estimation of the intensity of an inhomogenous Poisson process, among other counting processes, or the estimation of the mean of a random vector with nonnegative components.     

Discretized likelihood methods—asymptotic properties of discretized likelihood estimators (DLE's)

Suppose thatX ₁,X ₂, ...,X _n , ... is a sequence of i.i.d. random variables with a densityf(x, θ). Letc _n be a maximum order of consistency. We consider a solution \(\hat \theta _n \) of the discretized likelihood equation $$\sum\limits_{i = 1}^n {\log f(X_i ,\hat \theta _n + rc_n^{ - 1} ) - } \sum\limits_{i = 1}^n {\log f(X_i ,\hat \theta _n ) = a_n (\hat \theta _n ,r)} $$ wherea _n (θ,r) is chosen so that \(\hat \theta _n \) is asymptotically median unbiased (AMU). Then the solution \(\hat \theta _n \) is called a discretized likelihood estimator (DLE). In this paper it is shown in comparison with DLE that a maximum likelihood estimator (MLE) is second order asymptotically efficient but not third order asymptotically efficient in the regular case. Further it is seen that the asymptotic efficiency (including higher order cases) may be systematically discussed by the discretized likelihood methods.     

Exact and approximate EM estimation of mutually exciting hawkes processes

Motivated by the availability of continuous event sequences that trace the social behavior in a population e.g. email, we believe that mutually exciting Hawkes processes provide a realistic and informative model for these sequences. For complex mutually exciting processes, the numerical optimization used for univariate self exciting processes may not provide stable estimates. Furthermore, convergence can be exceedingly slow, making estimation computationally expensive and multiple random restarts doubly so. We derive an expectation maximization algorithm for maximum likelihood estimation mutually exciting processes that is faster, more robust, and less biased than estimation based on numerical optimization. For an exponentially decaying excitement function, each EM step can be computed in a single $O(N)$ pass through the data, for $N$ observations, without requiring the entire dataset to be in memory. More generally, exact inference is $\Theta (N^{2})$ , but we identify some simple $\Theta (N)$ approximation strategies that seem to provide good estimates while reducing the computational cost.     

Quasi-factors in ergodic theory

Motivated by the notion of quasi-factor in topological dynamics, we introduce an analogous notion in the context of ergodic theory. For two processes,X andY , we haveX?Y if and only ifY has a factor which is isomorphic to a quasi-factor ofX. On the other hand, weakly mixing processes can have nontrivial quasifactors which are not w.m. We characterize those ergodic processes which admit only trivial continuous ergodic quasi-factors, and use this characterization to conclude that a process with minimal selfjoinings is of this type. From this we derive the fact that for every suchX and any ergodicY eitherX ⊥Y orY extends some symmetric product ofX.     

Latent class CUB models

The paper proposes a latent class version of Combination of Uniform and (shifted) Binomial random variables ( CUB ) models for ordinal data to account for unobserved heterogeneity. The extension, called  LC-CUB , is useful when the heterogeneity is originated by clusters of respondents not identified by covariates: this may generate a multimodal response distribution, which cannot be adequately described by a standard  CUB model. The  LC-CUB model is a finite mixture of  CUB models yielding a multimodal theoretical distribution. Model identification is achieved by constraining the uncertainty parameters to be constant across latent classes. A simulation experiment shows the performance of the maximum likelihood estimator, whereas the usefulness of the approach is illustrated by means of a case study on political self-placement measured on an ordinal scale.     

On inference for fractional differential equations

Based on Malliavin calculus tools and approximation results, we show how to compute a maximum likelihood type estimator for a rather general differential equation driven by a fractional Brownian motion with Hurst parameter $H>1/2$ . Rates of convergence for the approximation task are provided, and numerical experiments show that our procedure leads to good results in terms of estimation.     

Generalized sub-Gaussian fractional Brownian motion queueing model

It is well known that often the one-dimensional distribution of a queue content is not Gaussian but its tails behave like a Gaussian. We propose to consider a general class of processes, namely the class of $\varphi $ -sub-Gaussian random processes, which is more general than the Gaussian one and includes non-Gaussian processes. The class of sub-Gaussian random processes contains Gaussian processes also and therefore is of special interest. In this paper we provide an estimate for the queue content distribution of a fluid queue fed by $N$ independent strictly $\varphi $ -sub-Gaussian generalized fractional Brownian motion input processes. We obtain an upper estimate of buffer overflow probability in a finite buffer system defined on any finite time interval $[a,b]$ or infinite interval $[0,\infty )$ . The derived estimate captures more accurately the performance of the queueing system for a wider-range of input processes.     

Minimal simplices inscribed in a convex body

Measuring how far a convex body $\mathcal{K }$ (of dimension $n$ ) with a base point ${O}\in \,\text{ int }\,\mathcal{K }$ is from an inscribed simplex $\Delta \ni {O}$ in “minimal” position, the interior point ${O}$ can display regular or singular behavior. If ${O}$ is a regular point then the $n+1$ chords emanating from the vertices of $\Delta $ and meeting at ${O}$ are affine diameters, chords ending in pairs of parallel hyperplanes supporting $\mathcal{K }$ . At a singular point ${O}$ the minimal simplex $\Delta $ degenerates. In general, singular points tend to cluster near the boundary of $\mathcal{K }$ . As connection to a number of difficult and unsolved problems about affine diameters shows, regular points are elusive, often non-existent. The first result of this paper uses Klee’s fundamental inequality for the critical ratio and the dimension of the critical set to obtain a general existence for regular points in a convex body with large distortion (Theorem A). This, in various specific settings, gives information about the structure of the set of regular and singular points (Theorem B). At the other extreme when regular points are in abundance, a detailed study of examples leads to the conjecture that the simplices are the only convex bodies with no singular points. The second and main result of this paper is to prove this conjecture in two different settings, when (1) $\mathcal{K }$ has a flat point on its boundary, or (2) $\mathcal{K }$ has $n$ isolated extremal points (Theorem C).     

Characterization of extendible distributions with exponential minima via processes that are infinitely divisible with respect to time

We present a stochastic representation for multivariate extendible distributions with exponential minima (exEM), whose components are conditionally iid in the sense of de Finetti’s theorem. It is shown that the “exponential minima property” is in one-to-one correspondence with the conditional cumulative hazard rate process being infinitely divisible with respect to time (IDT). The Laplace exponents of non-decreasing IDT processes are given in terms of a Bernstein function applied to the state space variable and are linear in time. Examples for IDT processes comprise killed Lévy subordinators, monomials whose slope is randomized by a stable random variable, and several combinations thereof. As a byproduct of our results, we provide an alternative proof (and a mild generalization) of the important conclusion in Genest and Rivest (Stat. Probab. Lett. 8:207211, 1989), stating that the only copula which is both Archimedean and of extreme-value kind is the Gumbel copula. Finally, we show that when the subfamily of strong IDT processes is used in the construction leading to exEM, the result is the proper subclass of extendible min-stable multivariate exponential (exMSMVE) distributions.     

A Question from a Famous Paper of Erdős

Given a convex body $K$ K , consider the smallest number $N$ N so that there is a point $P\in \partial K$ P ∈ ? K such that every circle centred at $P$ P intersects $\partial K$ ? K in at most $N$ N points. In 1946 Erd?s conjectured that $N=2$ N = 2 for all $K$ K , but there are convex bodies for which this is not the case. As far as we know there is no known global upper bound. We show that no convex body has $N=\infty $ N = ∞ and that there are convex bodies for which $N = 6$ N = 6 .     

Central limit theorems for empirical product densities of stationary point processes

We prove the asymptotic normality of kernel estimators of second- and higher-order product densities (and of the pair correlation function) for spatially homogeneous (and isotropic) point processes observed on a sampling window \(W_n\) , which is assumed to expand unboundedly in all directions as \(n \rightarrow \infty \,\) . We first study the asymptotic behavior of the covariances of the empirical product densities under minimal moment and weak dependence assumptions. The proof of the main results is based on the Brillinger-mixing property of the underlying point process and certain smoothness conditions on the higher-order reduced cumulant measures. Finally, the obtained limit theorems enable us to construct \(\chi ^2\) -goodness-of-fit tests for hypothetical product densities.     

On maximum likelihood estimation of the concentration parameter of von Mises–Fisher distributions

Maximum likelihood estimation of the concentration parameter of von Mises–Fisher distributions involves inverting the ratio \(R_\nu = I_{\nu +1} / I_\nu \) of modified Bessel functions and computational methods are required to invert these functions using approximative or iterative algorithms. In this paper we use Amos-type bounds for \(R_\nu \) to deduce sharper bounds for the inverse function, determine the approximation error of these bounds, and use these to propose a new approximation for which the error tends to zero when the inverse of \(R_\nu \) is evaluated at values tending to \(1\) (from the left). We show that previously introduced rational bounds for \(R_\nu \) which are invertible using quadratic equations cannot be used to improve these bounds.     

On adaptive minimax density estimation on R^d

We address the problem of adaptive minimax density estimation on \(\mathbb{R }^d\) with \(\mathbb{L }_p\) -loss on the anisotropic Nikol’skii classes. We fully characterize behavior of the minimax risk for different relationships between regularity parameters and norm indexes in definitions of the functional class and of the risk. In particular, we show that there are four different regimes with respect to the behavior of the minimax risk. We develop a single estimator which is (nearly) optimal in order over the complete scale of the anisotropic Nikol’skii classes. Our estimation procedure is based on a data-driven selection of an estimator from a fixed family of kernel estimators.