Automated selection of <Emphasis Type="Italic">r</Emphasis> for the <Emphasis Type="Italic">r</Emphasis> largest order statistics approach with adjustment for sequential testing

The r largest order statistics approach is widely used in extreme value analysis because it may use more information from the data than just the block maxima. In practice, the choice of r is critical. If r is too large, bias can occur; if too small, the variance of the estimator can be high. The limiting distribution of the r largest order statistics, denoted by GEV\(_r\), extends that of the block maxima. Two specification tests are proposed to select r sequentially. The first is a score test for the GEV\(_r\) distribution. Due to the special characteristics of the GEV\(_r\) distribution, the classical chi-square asymptotics cannot be used. The simplest approach is to use the parametric bootstrap, which is straightforward to implement but computationally expensive. An alternative fast weighted bootstrap or multiplier procedure is developed for computational efficiency. The second test uses the difference in estimated entropy between the GEV\(_r\) and GEV\(_{r-1}\) models, applied to the r largest order statistics and the \(r-1\) largest order statistics, respectively. The asymptotic distribution of the difference statistic is derived. In a large scale simulation study, both tests held their size and had substantial power to detect various misspecification schemes. A new approach to address the issue of multiple, sequential hypotheses testing is adapted to this setting to control the false discovery rate or familywise error rate. The utility of the procedures is demonstrated with extreme sea level and precipitation data.     

A discrepancy bound for deterministic acceptance-rejection samplers beyond $$N^{-1/2}$$ in dimension 1

In this paper we consider an acceptance-rejection (AR) sampler based on deterministic driver sequences. We prove that the discrepancy of an N element sample set generated in this way is bounded by \(\mathcal {O} (N^{-2/3}\log N)\), provided that the target density is twice continuously differentiable with non-vanishing curvature and the AR sampler uses the driver sequence \(\mathcal {K}_M= \{( j \alpha , j \beta ) ~~ mod~~1 \mid j = 1,\ldots ,M\},\) where \(\alpha ,\beta \) are real algebraic numbers such that \(1,\alpha ,\beta \) is a basis of a number field over \(\mathbb {Q}\) of degree 3. For the driver sequence \(\mathcal {F}_k= \{ ({j}/{F_k}, \{{jF_{k-1}}/{F_k}\} ) \mid j=1,\ldots , F_k\},\) where \(F_k\) is the k-th Fibonacci number and \(\{x\}=x-\lfloor x \rfloor \) is the fractional part of a non-negative real number x, we can remove the \(\log \) factor to improve the convergence rate to \(\mathcal {O}(N^{-2/3})\), where again N is the number of samples we accepted. We also introduce a criterion for measuring the goodness of driver sequences. The proposed approach is numerically tested by calculating the star-discrepancy of samples generated for some target densities using \(\mathcal {K}_M\) and \(\mathcal {F}_k\) as driver sequences. These results confirm that achieving a convergence rate beyond \(N^{-1/2}\) is possible in practice using \(\mathcal {K}_M\) and \(\mathcal {F}_k\) as driver sequences in the acceptance-rejection sampler.     

Random projections for Bayesian regression

This article deals with random projections applied as a data reduction technique for Bayesian regression analysis. We show sufficient conditions under which the entire d-dimensional distribution is approximately preserved under random projections by reducing the number of data points from n to \(k\in O({\text {poly}}(d/\varepsilon ))\) in the case \(n\gg d\). Under mild assumptions, we prove that evaluating a Gaussian likelihood function based on the projected data instead of the original data yields a \((1+O(\varepsilon ))\)-approximation in terms of the \(\ell _2\) Wasserstein distance. Our main result shows that the posterior distribution of Bayesian linear regression is approximated up to a small error depending on only an \(\varepsilon \)-fraction of its defining parameters. This holds when using arbitrary Gaussian priors or the degenerate case of uniform distributions over \(\mathbb {R}^d\) for \(\beta \). Our empirical evaluations involve different simulated settings of Bayesian linear regression. Our experiments underline that the proposed method is able to recover the regression model up to small error while considerably reducing the total running time.     

Investigation of the widely applicable Bayesian information criterion

The widely applicable Bayesian information criterion (WBIC) is a simple and fast approximation to the model evidence that has received little practical consideration. WBIC uses the fact that the log evidence can be written as an expectation, with respect to a powered posterior proportional to the likelihood raised to a power \(t^*\in {(0,1)}\), of the log deviance. Finding this temperature value \(t^*\) is generally an intractable problem. We find that for a particular tractable statistical model that the mean squared error of an optimally-tuned version of WBIC with correct temperature \(t^*\) is lower than an optimally-tuned version of thermodynamic integration (power posteriors). However in practice WBIC uses the a canonical choice of \(t=1/\log (n)\). Here we investigate the performance of WBIC in practice, for a range of statistical models, both regular models and singular models such as latent variable models or those with a hierarchical structure for which BIC cannot provide an adequate solution. Our findings are that, generally WBIC performs adequately when one uses informative priors, but it can systematically overestimate the evidence, particularly for small sample sizes.     

Rate of uniform consistency for a class of mode regression on functional stationary ergodic data

The aim of this paper is to study the asymptotic properties of a class of kernel conditional mode estimates whenever functional stationary ergodic data are considered. To be more precise on the matter, in the ergodic data setting, we consider a random elements (X, Z) taking values in some semi-metric abstract space \(E\times F\). For a real function \(\varphi \) defined on the space F and \(x\in E\), we consider the conditional mode of the real random variable \(\varphi (Z)\) given the event “\(X=x\)”. While estimating the conditional mode function, say \(\theta _\varphi (x)\), using the well-known kernel estimator, we establish the strong consistency with rate of this estimate uniformly over Vapnik–Chervonenkis classes of functions \(\varphi \). Notice that the ergodic setting offers a more general framework than the usual mixing structure. Two applications to energy data are provided to illustrate some examples of the proposed approach in time series forecasting framework. The first one consists in forecasting the daily peak of electricity demand in France (measured in Giga-Watt). Whereas the second one deals with the short-term forecasting of the electrical energy (measured in Giga-Watt per Hour) that may be consumed over some time intervals that cover the peak demand.     

Point process-based Monte Carlo estimation

This paper addresses the issue of estimating the expectation of a real-valued random variable of the form \(X = g(\mathbf {U})\) where g is a deterministic function and \(\mathbf {U}\) can be a random finite- or infinite-dimensional vector. Using recent results on rare event simulation, we propose a unified framework for dealing with both probability and mean estimation for such random variables, i.e. linking algorithms such as Tootsie Pop Algorithm or Last Particle Algorithm with nested sampling. Especially, it extends nested sampling as follows: first the random variable X does not need to be bounded any more: it gives the principle of an ideal estimator with an infinite number of terms that is unbiased and always better than a classical Monte Carlo estimator—in particular it has a finite variance as soon as there exists \(k \in \mathbb {R}> 1\) such that \({\text {E}}\left[ X^k \right] < \infty \). Moreover we address the issue of nested sampling termination and show that a random truncation of the sum can preserve unbiasedness while increasing the variance only by a factor up to 2 compared to the ideal case. We also build an unbiased estimator with fixed computational budget which supports a Central Limit Theorem and discuss parallel implementation of nested sampling, which can dramatically reduce its running time. Finally we extensively study the case where X is heavy-tailed.     

Multiplier bootstrap methods for conditional distributions

The multiplier bootstrap is a fast and easy-to-implement alternative to the standard bootstrap; it has been used successfully in many statistical contexts. In this paper, resampling methods based on multipliers are proposed in a general framework where one investigates the stochastic behavior of a random vector \(\mathbf {Y}\in \mathbb {R}^d\) conditional on a covariate \(X \in \mathbb {R}\). Specifically, two versions of the multiplier bootstrap adapted to empirical conditional distributions are introduced as alternatives to the conditional bootstrap and their asymptotic validity is formally established. As the method walks hand-in-hand with the functional delta method, theory around the estimation of statistical functionals is developed accordingly; this includes the interval estimation of conditional mean and variance, conditional correlation coefficient, Kendall’s dependence measure and copula. Composite inference about univariate and joint conditional distributions is also considered. The sample behavior of the new bootstrap schemes and related estimation methods are investigated via simulations and an illustration on real data is provided.     

Semi-parametric bivariate polychotomous ordinal regression

A pair of polychotomous random variables \((Y_1,Y_2)^\top =:{\varvec{Y}}\), where each \(Y_j\) has a totally ordered support, is studied within a penalized generalized linear model framework. We deal with a triangular generating process for \({\varvec{Y}}\), a structure that has been employed in the literature to control for the presence of residual confounding. Differently from previous works, however, the proposed model allows for a semi-parametric estimation of the covariate-response relationships. In this way, the risk of model mis-specification stemming from the imposition of fixed-order polynomial functional forms is also reduced. The proposed estimation methods and related inferential results are finally applied to study the effect of education on alcohol consumption among young adults in the UK.     

On the $$L_p$$ norms of kernel regression estimators for incomplete data with applications to classification

We consider kernel methods to construct nonparametric estimators of a regression function based on incomplete data. To tackle the presence of incomplete covariates, we employ Horvitz–Thompson-type inverse weighting techniques, where the weights are the selection probabilities. The unknown selection probabilities are themselves estimated using (1) kernel regression, when the functional form of these probabilities are completely unknown, and (2) the least-squares method, when the selection probabilities belong to a known class of candidate functions. To assess the overall performance of the proposed estimators, we establish exponential upper bounds on the \(L_p\) norms, \(1\le p<\infty \), of our estimators; these bounds immediately yield various strong convergence results. We also apply our results to deal with the important problem of statistical classification with partially observed covariates.     

A new universal resample-stable bootstrap-based stopping criterion for PLS component construction

We develop a new robust stopping criterion for partial least squares regression (PLSR) component construction, characterized by a high level of stability. This new criterion is universal since it is suitable both for PLSR and extensions to generalized linear regression (PLSGLR). The criterion is based on a non-parametric bootstrap technique and must be computed algorithmically. It allows the testing of each successive component at a preset significance level \(\alpha \). In order to assess its performance and robustness with respect to various noise levels, we perform dataset simulations in which there is a preset and known number of components. These simulations are carried out for datasets characterized both by \(n>p\), with n the number of subjects and p the number of covariates, as well as for \(n<p\). We then use t-tests to compare the predictive performance of our approach with other common criteria. The stability property is in particular tested through re-sampling processes on a real allelotyping dataset. An important additional conclusion is that this new criterion gives globally better predictive performances than existing ones in both the PLSR and PLSGLR (logistic and poisson) frameworks.     

Transformation approaches of linear random-effects models

Assume that a linear random-effects model \(\mathbf{y}= \mathbf{X}\varvec{\beta }+ \varvec{\varepsilon }= \mathbf{X}(\mathbf{A}\varvec{\alpha }+ \varvec{\gamma }) + \varvec{\varepsilon }\) is transformed as \(\mathbf{T}\mathbf{y}= \mathbf{T}\mathbf{X}\varvec{\beta }+ \mathbf{T}\varvec{\varepsilon }= \mathbf{T}\mathbf{X}(\mathbf{A}\varvec{\alpha }+ \varvec{\gamma }) + \mathbf{T}\varvec{\varepsilon }\) by pre-multiplying a given matrix \(\mathbf{T}\) of arbitrary rank. These two models are not necessarily equivalent unless \(\mathbf{T}\) is of full column rank, and we have to work with this derived model in many situations. Because predictors/estimators of the parameter spaces under the two models are not necessarily the same, it is primary work to compare predictors/estimators in the two models and to establish possible links between the inference results obtained from two models. This paper presents a general algebraic approach to the problem of comparing best linear unbiased predictors (BLUPs) of parameter spaces in an original linear random-effects model and its transformations, and provides a group of fundamental and comprehensive results on mathematical and statistical properties of the BLUPs. In particular, we construct many equalities for the BLUPs under an original linear random-effects model and its transformations, and obtain necessary and sufficient conditions for the equalities to hold.     

A standardized distance-based index to assess the quality of space-filling designs

One of the most used criterion for evaluating space-filling design in computer experiments is the minimal distance between pairs of points. The focus of this paper is to propose a normalized quality index that is based on the distribution of the minimal distance when points are drawn independently from the uniform distribution over the unit hypercube. Expressions of this index are explicitly given in terms of polynomials under any \(L_p\) distance. When the size of the design or the dimension of the space is large, approximations relying on extreme value theory are derived. Some illustrations of our index are presented on simulated data and on a real problem.     

Wavelet regression estimations with strong mixing data

Using a wavelet basis, we establish in this paper upper bounds of wavelet estimation on \( L^{p}({\mathbb {R}}^{d}) \) risk of regression functions with strong mixing data for \( 1\le p<\infty \). In contrast to the independent case, these upper bounds have different analytic formulae for \(p\in [1, 2]\) and \(p\in (2, +\infty )\). For \(p=2\), it turns out that our result reduces to a theorem of Chaubey et al. (J Nonparametr Stat 25:53–71, 2013); and for \(d=1\) and \(p=2\), it becomes the corresponding theorem of Chaubey and Shirazi (Commun Stat Theory Methods 44:885–899, 2015).     

Hypotheses testing about the drift parameter in linear stochastic differential equation driven by stable processes

In this paper, we consider the problem of hypotheses testing about the drift parameter \(\theta \) in the process \(\text {d}Y^{\delta }_{t} = \theta \dot{f}(t)Y^{\delta }_{t}\text {d}t + b(t)\text {d}L^{\delta }_{t}\) driven by symmetric \(\delta \)-stable Lévy process \(L^{\delta }_{t}\) with \(\dot{f}(t)\) being the derivative of a known increasing function f(t) and b(t) being known as well. We consider the hypotheses testing \(H_{0}: \theta \le 0\) and \(K_{0}: \theta =0\) against the alternatives \(H_{1}: \theta >0\) and \(K_{1}: \theta \ne 0\), respectively. For these hypotheses, we propose inverse methods, which are motivated by sequential approach, based on the first hitting time of the observed process (or its absolute value) to a pre-specified boundary or two boundaries until some given time. The applicability of these methods is illustrated. For the case \(Y^{\delta }_{0}=0\), we are able to calculate the values of boundaries and finite observed times more directly. We are able to show the consistencies of proposed tests for \(Y^{\delta }_{0}\ge 0\) with \(\delta \in (1,2]\) and for \(Y^{\delta }_{0}=0\) with \(\delta \in (0,2]\) under quite mild conditions.     

A coordinate descent algorithm for computing penalized smooth quantile regression

The computation of penalized quantile regression estimates is often computationally intensive in high dimensions. In this paper we propose a coordinate descent algorithm for computing the penalized smooth quantile regression (cdaSQR) with convex and nonconvex penalties. The cdaSQR approach is based on the approximation of the objective check function, which is not differentiable at zero, by a modified check function which is differentiable at zero. Then, using the maximization-minimization trick of the gcdnet algorithm (Yang and Zou in, J Comput Graph Stat 22(2):396–415, 2013), we update each coefficient simply and efficiently. In our implementation, we consider the convex penalties \(\ell _1+\ell _2\) and the nonconvex penalties SCAD (or MCP) \(+ \ell _2\). We establishe the convergence property of the csdSQR with \(\ell _1+\ell _2\) penalty. The numerical results show that our implementation is an order of magnitude faster than its competitors. Using simulations we compare the speed of our algorithm to its competitors. Finally, the performance of our algorithm is illustrated on three real data sets from diabetes, leukemia and Bardet–Bidel syndrome gene expression studies.     

Fast parallel $$\alpha $$-stable distribution function evaluation and parameter estimation using OpenCL in GPGPUs

\(\alpha \)-Stable distributions are a family of probability distributions found to be suitable to model many complex processes and phenomena in several research fields, such as medicine, physics, finance and networking, among others. However, the lack of closed expressions makes their evaluation analytically intractable, and alternative approaches are computationally expensive. Existing numerical programs are not fast enough for certain applications and do not make use of the parallel power of general purpose graphic processing units. In this paper, we develop novel parallel algorithms for the probability density function and cumulative distribution function—including a parallel Gauss–Kronrod quadrature—, quantile function, random number generator and maximum likelihood estimation of \(\alpha \)-stable distributions using OpenCL, achieving significant speedups and precision in all cases. Thanks to the use of OpenCL, we also evaluate the results of our library with different GPU architectures.     

Stochastically optimal bootstrap sample size for shrinkage-type statistics

In nonregular problems where the conventional \(n\) out of \(n\) bootstrap is inconsistent, the \(m\) out of \(n\) bootstrap provides a useful remedy to restore consistency. Conventionally, optimal choice of the bootstrap sample size \(m\) is taken to be the minimiser of a frequentist error measure, estimation of which has posed a major difficulty hindering practical application of the \(m\) out of \(n\) bootstrap method. Relatively little attention has been paid to a stronger, stochastic, version of the optimal bootstrap sample size, defined as the minimiser of an error measure calculated directly from the observed sample. Motivated by this stronger notion of optimality, we develop procedures for calculating the stochastically optimal value of \(m\). Our procedures are shown to work under special forms of Edgeworth-type expansions which are typically satisfied by statistics of the shrinkage type. Theoretical and empirical properties of our methods are illustrated with three examples, namely the James–Stein estimator, the ridge regression estimator and the post-model-selection regression estimator.     

The expected bit complexity of the von Neumann rejection algorithm

In 1952, von Neumann introduced the rejection method for random variate generation. We revisit this algorithm when we have a source of perfect bits at our disposal. In this random bit model, there are universal lower bounds for generating a random variate with a given density to within an accuracy \(\epsilon \) derived by Knuth and Yao, and refined by the authors. In general, von Neumann’s method fails in this model. We propose a modification that insures proper behavior for all Riemann-integrable densities on compact sets, and show that the expected number of random bits needed behaves optimally with respect to universal lower bounds. In particular, we introduce the notion of an oracle that evaluates the supremum and infimum of a function on any rectangle of \({\mathbb {R}}^{d}\), and develop a quadtree-style extension of the classical rejection method.     

On the decomposition by sources of the Zenga 1984 point and synthetic inequality indexes

In this work we provide a decomposition by sources of the inequality index \(\zeta \) defined by Zenga (Giornale degli Economisti e Annali di economia 43(5–6):301–326, 1984). The source contributions are obtained with the method proposed in Zenga et al. (Stat Appl X(1):3–31, 2012) and Zenga (Stat Appl XI(2):133–161, 2013), that allows to compare different inequality measures. This method is based on the decomposition of inequality curves. To apply this decomposition to the index \(\zeta \) and its inequality curve, we adapt the method to the “cograduation” table. Moreover, we consider the case of linear transformation of sources and analyse the corresponding results.