Robust and Efficient Estimation for the Generalized Pareto Distribution

In this article we implement the minimum density power divergence estimator (MDPDE) for the shape and scale parameters of the generalized Pareto distribution (GPD). The MDPDE is indexed by a constant 0 that controls the trade-off between robustness and efficiency. As increases, robustness increases and efficiency decreases. For = 0 the MDPDE is equivalent to the maximum likelihood estimator (MLE). We show that for > 0 the MDPDE for the GPD has a bounded influence function. For < 0.2 the MDPDE maintains good asymptotic relative efficiencies, usually above 90%. The results from a Monte Carlo study agree with these asymptotic calculations. The MDPDE is asymptotically normally distributed if the shape parameter is less than (1 + )/(2 + ), and estimators for standard errors are readily computed under this restriction. We compare the MDPDE, MLE, Dupuis optimally-biased robust estimator (OBRE), and Peng and Welshs Medians estimator for the parameters. The simulations indicate that the MLE has the highest efficiency under uncontaminated GPDs. However, for the GPD contaminated with gross errors OBRE and MDPDE are more efficient than the MLE. For all the simulated models that we studied the Medians estimator had poor performance.AMS 2000 Subject Classification. Primary—62F35, Secondary—62G35     

On maximum likelihood estimation of the drift matrix of a degenerated O–U process

In this work, we consider a 2n-dimension Ornstein–Uhlenbeck (O–U) process with a singular diffusion matrix. This process represents a currently used model for mechanical systems subject to random vibrations. We study the problem of estimating the drift parameters of the stochastic differential equation that governs the O–U process. The maximum likelihood estimator proposed and explored in Koncz (J Anal Math 13(1):75–91, 1987) is revisited and applied to our model. We prove the local asymptotic normality property and the convergence of moments of the estimator. Simulation studies based on representative examples taken from the literature illustrate the obtained theoretical results.     

HAC estimation and strong linearity testing in weak ARMA models

In the framework of ARMA models, we consider testing the reliability of the standard asymptotic covariance matrix (ACM) of the least-squares estimator. The standard formula for this ACM is derived under the assumption that the errors are independent and identically distributed, and is in general invalid when the errors are only uncorrelated. The test statistic is based on the difference between a conventional estimator of the ACM of the least-squares estimator of the ARMA coefficients and its robust HAC-type version. The asymptotic distribution of the HAC estimator is established under the null hypothesis of independence, and under a large class of alternatives. The asymptotic distribution of the proposed statistic is shown to be a standard χ² under the null, and a noncentral χ² under the alternatives. The choice of the HAC estimator is discussed through asymptotic power comparisons. The finite sample properties of the test are analyzed via Monte Carlo simulation.     

Robust residual‐ and recovery‐based a posteriori error estimators for a multipoint flux mixed finite element discretization of interface problems

We consider a posteriori error estimation for a multipoint flux mixed finite element method for two‐dimensional elliptic interface problems. Within the class of modified quasi‐monotonically distributed coefficients, we derive a residual‐type a posteriori error estimator of the weighted sum of the scalar and flux errors which is robust with respect to the jumps of the coefficients. Moreover, we develop robust implicit and explicit recovery‐type estimators through gradient recovery in an H(curl)‐conforming finite element space. In particular, we apply a modified L² projection in the implicit recovery procedure so as to reduce the computational cost of the recovered gradient. Numerical experiments confirm the theoretical results.     

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.     

A posteriori error analysis of multipoint flux mixed finite element methods for interface problems

In this paper, the multipoint flux mixed finite element method is used to approximate the flux of two-dimensional elliptic interface problems. Within the class of modified quasi-monotonically distributed coefficients, we derive uniformly robust residual-type a posteriori error estimators for the flux error. Based on the residual-type estimator, we further develop robust implicit and explicit recovery-type estimators through gradient recovery in H(curl) conforming finite element spaces. Numerical experiments are presented to support the theoretical results.     

Robust regression function estimation

A robust estimator of the regression function is proposed combining kernel methods as introduced for density estimation and robust location estimation techniques. Weak and strong consistency and asymptotic normality are shown under mild conditions on the kernel sequence. The asymptotic variance is a product from a factor depending only on the kernel and a factor similar to the asymptotic variance in robust estimation of location. The estimation is minimax robust in the sense of Huber (1964). Robust estimation of a location parameter. Ann. Math. Statist.33 73–101.     

Estimateur hiérarchique robuste pour un problème de perturbation singulière

In this Note, we show that a modified and simplified version of the estimator of Bank–Weiser can be used to define a robust a posteriori error estimator for singularly perturbed problem. We prove without comparison with a residual estimator or saturation assumption, the equivalence of the estimator with the error in the energy norm and the robusteness with respect to the diffusion coefficient. To cite this article: B. Achchab et al., C. R. Acad. Sci. Paris, Ser. I 336 (2003).     

Preliminary test estimation in system regression models in view of asymmetry

In this paper, we consider the system regression model introduced by Arashi and Roozbeh (Comput Stat 30:359–376, 2015) and study the performance of the feasible preliminary test estimator (FPTE) both analytically and computationally, under the assumption that constraints may hold on the vector parameter space. The performance of the FPTE is analysed through a Monte Carlo simulation study under bounded and or asymmetric loss functions. An application of the so-called Cobb–Douglas production function in economic modelling together with the results from the simulation study shows that the bounded linear exponential (BLINEX) loss function outperforms the linear exponential loss function (LINEX) by comparing risk values.     

Estimation of a tail index based on minimum density power divergence

In this paper, we consider the minimum density power divergence estimator for the tail index of heavy tailed distributions in strong mixing processes. It is shown that the estimator is consistent and asymptotically normal under regularity conditions. The simulation results demonstrate that the estimator is robust in the presence of outliers.     

Estimating the Gerber–Shiu function in the perturbed compound Poisson model by Laguerre series expansion

In this paper, we study the statistical estimation of the Gerber–Shiu function in the compound Poisson risk model perturbed by diffusion. This problem has been solved in [32] by the Fourier–Sinc series expansion method. Different from [32], we use the Laguerre series to expand the Gerber–Shiu function and propose a relevant estimator. The estimator is easily computed and has fast convergence rate. Various simulation studies are presented to confirm that the estimator performs well when the sample size is finite.     

Coupling Importance Sampling and Multilevel Monte Carlo using Sample Average Approximation

In this work, we propose a smart idea to couple importance sampling and Multilevel Monte Carlo (MLMC). We advocate a per level approach with as many importance sampling parameters as the number of levels, which enables us to handle the different levels independently. The search for parameters is carried out using sample average approximation, which basically consists in applying deterministic optimisation techniques to a Monte Carlo approximation rather than resorting to stochastic approximation. Our innovative estimator leads to a robust and efficient procedure reducing both the discretization error (the bias) and the variance for a given computational effort. In the setting of discretized diffusions, we prove that our estimator satisfies a strong law of large numbers and a central limit theorem with optimal limiting variance, in the sense that this is the variance achieved by the best importance sampling measure (among the class of changes we consider), which is however non tractable. Finally, we illustrate the efficiency of our method on several numerical challenges coming from quantitative finance and show that it outperforms the standard MLMC estimator.     

Estimating and Visualizing Conditional Densities

Abstract

We consider the kernel estimator of conditional density and derive its asymptotic bias, variance, and mean-square error. Optimal bandwidths (with respect to integrated mean-square error) are found and it is shown that the convergence rate of the density estimator is order n ^–2/3. We also note that the conditional mean function obtained from the estimator is equivalent to a kernel smoother. Given the undesirable bias properties of kernel smoothers, we seek a modified conditional density estimator that has mean equivalent to some other nonparametric regression smoother with better bias properties. It is also shown that our modified estimator has smaller mean square error than the standard estimator in some commonly occurring situations. Finally, three graphical methods for visualizing conditional density estimators are discussed and applied to a data set consisting of maximum daily temperatures in Melbourne, Australia.     

Robust estimation of the Pickands dependence function under random right censoring

We consider robust nonparametric estimation of the Pickands dependence function under random right censoring. The estimator is obtained by applying the minimum density power divergence criterion to properly transformed bivariate observations. The asymptotic properties are investigated by making use of results for Kaplan–Meier integrals. We investigate the finite sample properties of the proposed estimator with a simulation experiment and illustrate its practical applicability on a dataset of insurance indemnity losses.     

Robust kernel-based regression with bounded influence for outliers

The kernel-based regression (KBR) method, such as support vector machine for regression (SVR) is a well-established methodology for estimating the nonlinear functional relationship between the response variable and predictor variables. KBR methods can be very sensitive to influential observations that in turn have a noticeable impact on the model coefficients. The robustness of KBR methods has recently been the subject of wide-scale investigations with the aim of obtaining a regression estimator insensitive to outlying observations. However, existing robust KBR (RKBR) methods only consider Y-space outliers and, consequently, are sensitive to X-space outliers. As a result, even a single anomalous outlying observation in X-space may greatly affect the estimator. In order to resolve this issue, we propose a new RKBR method that gives reliable result even if a training data set is contaminated with both Y-space and X-space outliers. The proposed method utilizes a weighting scheme based on the hat matrix that resembles the generalized M-estimator (GM-estimator) of conventional robust linear analysis. The diagonal elements of hat matrix in kernel-induced feature space are used as leverage measures to downweight the effects of potential X-space outliers. We show that the kernelized hat diagonal elements can be obtained via eigen decomposition of the kernel matrix. The regularized version of kernelized hat diagonal elements is also proposed to deal with the case of the kernel matrix having full rank where the kernelized hat diagonal elements are not suitable for leverage. We have shown that two kernelized leverage measures, namely, the kernel hat diagonal element and the regularized one, are related to statistical distance measures in the feature space. We also develop an efficiently kernelized training algorithm for the parameter estimation based on iteratively reweighted least squares (IRLS) method. The experimental results from simulated examples and real data sets demonstrate the robustness of our proposed method compared with conventional approaches.     

Combining the data from two normal populations to estimate the mean of one when their means difference is bounded

In this paper we address the problem of estimating θ₁ when , are observed and |θ₁−θ₂|?c for a known constant c. Clearly Y₂ contains information about θ₁. We show how the so-called weighted likelihood function may be used to generate a class of estimators that exploit that information. We discuss how the weights in the weighted likelihood may be selected to successfully trade bias for precision and thus use the information effectively. In particular, we consider adaptively weighted likelihood estimators where the weights are selected using the data. One approach selects such weights in accord with Akaike's entropy maximization criterion. We describe several estimators obtained in this way. However, the maximum likelihood estimator is investigated as a competitor to these estimators along with a Bayes estimator, a class of robust Bayes estimators and (when c is sufficiently small), a minimax estimator. Moreover we will assess their properties both numerically and theoretically. Finally, we will see how all of these estimators may be viewed as adaptively weighted likelihood estimators. In fact, an over-riding theme of the paper is that the adaptively weighted likelihood method provides a powerful extension of its classical counterpart.     

Quantile regression of right-censored length-biased data using the Buckley–James-type method

Length-biased data are encountered frequently due to prevalent cohort sampling in follow-up studies. Quantile regression provides great flexibility for assessing covariate effects on survival time, and is a useful alternative to Cox’s proportional hazards model and the accelerated failure time (AFT) model for survival analysis. In this paper, we develop a Buckley–James-type estimator for right-censored length-biased data under a quantile regression model. The problem of informative right-censoring of length-biased data induced by prevalent cohort sampling must be handled. Following on from the generalization of the Buckley–James-type estimator under the AFT model proposed by Ning et al. (Biometrics 67:1369–1378, 2011), we propose a Buckley–James-type estimating equation for regression coefficients in the quantile regression model and develop an iterative algorithm to obtain the estimates. The resulting estimator is consistent and asymptotically normal. We evaluate the performance of the proposed estimator on finite samples using extensive simulation studies. Analysis of real data is presented to illustrate our proposed methodology.     

Composite quantile regression estimation for P-GARCH processes

We consider the periodic generalized autoregressive conditional heteroskedasticity(P-GARCH) process and propose a robust estimator by composite quantile regression. We study some useful properties about the P-GARCH model. Under some mild conditions, we establish the asymptotic results of proposed estimator.The Monte Carlo simulation is presented to assess the performance of proposed estimator. Numerical study results show that our proposed estimation outperforms other existing methods for heavy tailed distributions.The proposed methodology is also illustrated by Va R on stock price data.     

Test for parameter change based on the estimator minimizing density-based divergence measures

In this paper we consider the problem of testing for a parameter change based on the cusum test proposed by Leeet al. (2003,Scandinavian Journal of Statistics,30, 781–796). The cusum test statistic is constructed via employing the estimator minimizing density-based divergence measures. It is shown that under regularity conditions, the test statistic has the limiting distribution of the sup of standard Brownian bridge. Simulation results demonstrate that the cusum test is robust when outliers exist.     

On the approximability of adjustable robust convex optimization under uncertainty

In this paper, we consider adjustable robust versions of convex optimization problems with uncertain constraints and objectives and show that under fairly general assumptions, a static robust solution provides a good approximation for these adjustable robust problems. An adjustable robust optimization problem is usually intractable since it requires to compute a solution for all possible realizations of uncertain parameters, while an optimal static solution can be computed efficiently in most cases if the corresponding deterministic problem is tractable. The performance of the optimal static robust solution is related to a fundamental geometric property, namely, the symmetry of the uncertainty set. Our work allows for the constraint and objective function coefficients to be uncertain and for the constraints and objective functions to be convex, thereby providing significant extensions of the results in Bertsimas and Goyal (Math Oper Res 35:284–305, 2010) and Bertsimas et al. (Math Oper Res 36: 24–54, 2011b) where only linear objective and linear constraints were considered. The models in this paper encompass a wide variety of problems in revenue management, resource allocation under uncertainty, scheduling problems with uncertain processing times, semidefinite optimization among many others. To the best of our knowledge, these are the first approximation bounds for adjustable robust convex optimization problems in such generality.