A constructive hypothesis test for the single-index models with two groups

Comparison of two-sample heteroscedastic single-index models, where both the scale and location functions are modeled as single-index models, is studied in this paper. We propose a test for checking the equality of single-index parameters when dimensions of covariates of the two samples are equal. Further, we propose two test statistics based on Kolmogorov–Smirnov and Cramér–von Mises type functionals. These statistics evaluate the difference of the empirical residual processes to test the equality of mean functions of two single-index models. Asymptotic distributions of estimators and test statistics are derived. The Kolmogorov–Smirnov and Cramér–von Mises test statistics can detect local alternatives that converge to the null hypothesis at a parametric convergence rate. To calculate the critical values of Kolmogorov–Smirnov and Cramér–von Mises test statistics, a bootstrap procedure is proposed. Simulation studies and an empirical study demonstrate the performance of the proposed procedures.     

Correcting statistical models via empirical distribution functions

We consider the two-sample homogeneity problem where the information contained in two samples is used to test the equality of the underlying distributions. In cases where one sample is simulated by a procedure modelling the data generating process of another observed sample, a mere rejection of the null hypothesis is unsatisfactory. Instead, the data analyst would like to know how the simulation can be improved. Based on the popular Kolmogorov–Smirnov test and a general mixture model, we propose an algorithm that determines an appropriate correction distribution function. Complementing the simulation sample by a given proportion of observations sampled from this distribution reduces the Kolmogorov–Smirnov distance between the modified and the observed sample. Therefore, the correction distribution indicates possible improvements to the current simulation process. We prove our algorithm to run in linear time when applied to sorted samples. We further illustrate its intuitive results on simulated as well as on real data sets from astrophysics and bioinformatics.     

Limit distributions and comparative asymptotic efficiency of the Smirnov — Kolmogorov statistics with a random index

Limit distributions for certain statistics of Smirnov — Kolmogorov type are obtained which consider the weak convergence of the corresponding empirical process. Approximate and precise asymptotic efficiencies of these statistics are computed. It is shown that they are worse in a certain sense than the classical Kolmogorov — Smirnov statistics.Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Instituta im. V. A. Steklova Akad. Nauk SSSR, Vol. 55, pp. 185–194, 1976.     

金融资产收益分布的混合高斯分析

本文研究了金融资产收益的混合高斯分布模型 ,给出了混合高斯分布的 Kolmogorov-Smirnov检验的方法 ,分析了金融资产收益的非高斯性及市场价格运动的有效性 .此外 ,用成分数目 K*、拟合误差 DK*n和主成分系数 p*k 描述金融资产收益的性质 ,对外汇银行同业拆借市场和中国股票市场实证分析     

Tests of symmetry for bivariate copulas

Tests are proposed for the hypothesis that the underlying copula of a continuous random pair is symmetric. The procedures are based on Cramér–von Mises and Kolmogorov–Smirnov functionals of a rank-based empirical process whose large-sample behaviour is obtained. The asymptotic validity of a re-sampling method to compute P values is also established. The technical arguments supporting the use of a Chi-squared test due to Jasson are also presented. A power study suggests that the proposed tests are more powerful than Jasson’s procedure under many scenarios of copula asymmetry. The methods are illustrated on a nutrient data set.     

Analogues of Classical Goodness-of-Fit Tests for Distribution Tails

Doklady Mathematics - We propose analogues of the classical Kolmogorov–Smirnov and omega-squared tests for goodness-of-fit testing of distribution tails. The consistency of the proposed tests...     

Hodges-Lehmann asymptotic efficiency of the Kolmogorov and Smirnov goodness-of-fit tests

One considers the Hodges-Lehmann asymptotic efficiency of the Kolmogorov and Smirnov goodness-of-fit tests, which measures the rate of the exponential decrease of the errors of the second kind, under a fixed significance level. It is shown that the Kolmogorov test is always asymptotically optimal in this sense, while the one-sided Smirnov test is asymptotically optimal under additional conditions imposed on the parametric family of distributions.Translated from Zapiski Nauchnykh Seminarov Leningradskogo Otdeleniya Matematicheskogo Instituta im. V. A. Steklova AN SSSR, Vol. 142, pp. 119–123, 1985.     

On large deviations of Kolmogorov-Smirnov-Renyi type statistics

One result of Smirnov's important paper [Uspehi Mat. Nauk.10, 179–206, (in Russian)] yields exponential bounds for the large deviations of his one-sided Smirnov statistic and the two-sided Kolmogorov statistic. In the present paper exponential bounds are given for the large deviations of a wide class of Kolmogorov-Smirnov-Renyi type statistics. As a by-product, exponential bounds for the large deviations of the corresponding limit distributions are obtained.     

Approximate Central Limit Theorems

We refine the classical Lindeberg–Feller central limit theorem by obtaining asymptotic bounds on the Kolmogorov distance, the Wasserstein distance, and the parameterized Prokhorov distances in terms of a Lindeberg index. We thus obtain more general approximate central limit theorems, which roughly state that the row-wise sums of a triangular array are approximately asymptotically normal if the array approximately satisfies Lindeberg’s condition. This allows us to continue to provide information in nonstandard settings in which the classical central limit theorem fails to hold. Stein’s method plays a key role in the development of this theory.     

Translated Poisson Approximation to Equilibrium Distributions of Markov Population Processes

The paper is concerned with the equilibrium distributions of continuous-time density dependent Markov processes on the integers. These distributions are known typically to be approximately normal, with \(O( 1 /{\sqrt{n}})\) error as measured in Kolmogorov distance. Here, an approximation in the much stronger total variation norm is established, without any loss in the asymptotic order of accuracy; the approximating distribution is a translated Poisson distribution having the same variance and (almost) the same mean. Our arguments are based on the Stein–Chen method and Dynkin’s formula.     

Probability distribution fitting of schedule overruns in construction projects

The probability of schedule overruns for construction and engineering projects can be ascertained using a ‘best fit’ probability distribution from an empirical distribution. The statistical characteristics of schedule overruns occurring in 276 Australian construction and engineering projects were analysed. Skewness and kurtosis values revealed that schedule overruns are non-Gaussian. Theoretical probability distributions were then fitted to the schedule overrun data; including the Kolmogorov–Smirnov, Anderson–Darling and Chi-Squared non-parametric tests to determine the ‘Goodness of Fit’. A Four Parameter Burr probability function best described the behaviour of schedule overruns, provided the best overall distribution fit and was used to calculate the probability of a schedule overrun being experienced. The statistical characteristics of contract size and schedule overruns were also analysed, and the Wakeby (<AU$1?m and AU$11–50?m), Three Parameter Log-logistic (AU$1–A$10?m) and Beta (AU$51–A$100?m and >AU$101?m) models provided the best distribution fits and were used to calculate schedule overrun probabilities by contract size.     

Kolmogorov–Smirnov‐type testing for the partial homogeneity of Markov processes—with application to credit risk

In banking, the default behaviour of the counterpart is not only of interest for the pricing of transactions under credit risk but also for the assessment of a portfolio credit risk. We develop a test against the hypothesis that default intensities are chronologically constant within a group of similar counterparts, e.g. a rating class. The Kolmogorov–Smirnov‐type test builds up on the asymptotic normality of counting processes in event history analysis. The right censoring accommodates for Markov processes with more than one no‐absorbing state. A simulation study and two examples of rating systems demonstrate that partial homogeneity can be assumed, however occasionally, certain migrations must be modelled and estimated inhomogeneously. Copyright © 2007 John Wiley & Sons, Ltd.     

PP multivariate random weighting method

According to the Projection Pursuit (PP) method and the random weighting method, we propose a PP random weighting method, and set up the asymptotic distribution theory and strong limit theorem of PP random weighting empirical process. Applying this method, we obtain two kinds of goodness-of-fit test for a multivariate distribution function, i.e., we get the random weighting approximations of PP Kolmogorov Smirnov statistics (PPKS) and PP Smirnov Cramér Von Mises statistics (PPSC), we prove that the asymptotic distribution of PPKS and PPSC are the same as those of their respective random weighting approximations.Supported by the National Natural Science Foundation of China.     

Good practice in retail credit scorecard assessment

In retail banking, predictive statistical models called ‘scorecards’ are used to assign customers to classes, and hence to appropriate actions or interventions. Such assignments are made on the basis of whether a customer's predicted score is above or below a given threshold. The predictive power of such scorecards gradually deteriorates over time, so that performance needs to be monitored. Common performance measures used in the retail banking sector include the Gini coefficient, the Kolmogorov–Smirnov statistic, the mean difference, and the information value. However, all of these measures use irrelevant information about the magnitude of scores, and fail to use crucial information relating to numbers misclassified. The result is that such measures can sometimes be seriously misleading, resulting in poor quality decisions being made, and mistaken actions being taken. The weaknesses of these measures are illustrated. Performance measures not subject to these risks are defined, and simple numerical illustrations are given.     

Normal approximation on Poisson spaces: Mehler’s formula,second order Poincaré inequalities and stabilization

We prove a new class of inequalities, yielding bounds for the normal approximation in the Wasserstein and the Kolmogorov distance of functionals of a general Poisson process (Poisson random measure). Our approach is based on an iteration of the classical Poincaré inequality, as well as on the use of Malliavin operators, of Stein’s method, and of an (integrated) Mehler’s formula, providing a representation of the Ornstein-Uhlenbeck semigroup in terms of thinned Poisson processes. Our estimates only involve first and second order difference operators, and have consequently a clear geometric interpretation. In particular we will show that our results are perfectly tailored to deal with the normal approximation of geometric functionals displaying a weak form of stabilization, and with non-linear functionals of Poisson shot-noise processes. We discuss two examples of stabilizing functionals in great detail: (i) the edge length of the k-nearest neighbour graph, (ii) intrinsic volumes of k-faces of Voronoi tessellations. In all these examples we obtain rates of convergence (in the Kolmogorov and the Wasserstein distance) that one can reasonably conjecture to be optimal, thus significantly improving previous findings in the literature. As a necessary step in our analysis, we also derive new lower bounds for variances of Poisson functionals.     

Discovering focal regions of slightly-aggregated sparse signals

The characteristic aspects of dynamic distortions on a lengthy time series of i.i.d. pure noise when embedded with slightly-aggregating sparse signals are summarized into a significantly shorter recurrence time process of a chosen extreme event. We first employ the Kolmogorov–Smirnov statistic to compare the empirical recurrence time distribution with the null geometry distribution when no signal being present in the original time series. The power of such a hypothesis testing depends on varying degrees of aggregation of sparse signals: from a completely random distribution of singletons to batches of various sizes on the entire temporal span. We demonstrate the Kolmogorov–Smirnov statistic capturing the dynamic distortions due to slightly-aggregating sparse signals better than does Tukey’s Higher Criticism statistic even when the batch size is as small as five. Secondly, after confirming the presence of signals in the pure noise time series, we apply the hierarchical factor segmentation (HFS) algorithm again based on the recurrence time process to compute focal segments that contain a significantly higher intensity of signals than do the rest of the temporal regions. In a computer experiment with a given fixed number of signals, the focal segments identified by the HFS algorithm afford many folds of signal intensity which also critically depend on the degree of aggregation of sparse signals. This ratio information can facilitate better sensitivity, equivalent to a smaller false discovery rate, if the signal-discovering protocol implemented within the computed focal regions is different from that used outside of the focal regions. We also numerically compute the specificity as the total number of signals contained in the computed collection of focal regions, which indicates the inherent difficulty in the task of sparse signal discovery.     

Nonparametric evaluation of the first passage time of degradation processes

This paper discusses a nonparametric method to approximate the first passage time (FPT) distribution of the degradation processes incorporating random effects if the process type is unknown. The FPT of a degradation process is unnecessarily observed since its density function can be approximated by inverting the empirical Laplace transform using the empirical saddlepoint method. The empirical Laplace transform is composed of the measured increments of the degradation processes. To evaluate the performance of the proposed method, the approximated FPT is compared with the theoretical FPT assuming a true underlying process. The nonparametric method discussed in this paper is shown to possess the comparatively small relative errors in the simulation study and performs well to capture the heterogeneity in the practical data analysis. To justify the fitting results, the goodness‐of‐fit tests including Kolmogorov‐Smirnov test and Cramér‐von Mises test are conducted, and subsequently, a bootstrap confidence interval is constructed in terms of the 90th percentile of the FPT distribution.     

Empirical processes with estimated parameters under auxiliary information

Empirical processes with estimated parameters are a well established subject in nonparametric statistics. In the classical theory they are based on the empirical distribution function which is the nonparametric maximum likelihood estimator for a completely unknown distribution function. In the presence of some “nonparametric” auxiliary information about the distribution, like a known mean or a known median, for example, the nonparametric maximum likelihood estimator is a modified empirical distribution function which puts random masses on the observations in order to take the available information into account [see Owen, Biometrika 75 (1988) 237–249, Ann. Statist. 18 (1990) 90–120, Empirical Likelihood, Chapman & Hall/CRC, London/Boca Raton, FL; Qin and Lawless, Ann. Statist. 22 (1994) 300–325]. Zhang [Metrika 46 (1997) 221–244] has proved a functional central limit theorem for the empirical process pertaining to this modified empirical distribution function. We will consider the corresponding empirical process with estimated parameters here and derive its asymptotic distribution. The limiting process is a centered Gaussian process with a complicated covariance function depending on the unknown parameter. The result becomes useful in practice through the bootstrap, which is shown to be consistent in case of a known mean. The performance of the resulting bootstrap goodness-of-fit test based on the Kolmogorov–Smirnov statistic is studied through simulations.     

The general form of the law of the iterated logarithm for generalized Kolmogorov-Smirnov-Renyi statistics

One result of Smirnov's important paper [Uspehi Mat. Nauk. 10, 179–206, (in Russian)] yields exponential bounds for the large deviations of his one-sided Smirnov statistic and the two-sided Kolmogorov statistic. In the present paper exponential bounds are given for the large deviations of a wide class of Kolmogorov-Smirnov-Renyi type statistics. As a by-product, exponential bounds for the large deviations of the corresponding limit distributions are obtained.     

随机估计和VDR检验

众所周知统计推断有三种理论:普遍承认的Neyman理论(频率学派),Bayes推断和信仰推断(Fiducial)。Bayes推断基于后验分布,由先验分布和样本分布求得。信仰推断是基于信仰分布(Confidence Distribution,简称CD),直接利用样本求得。两者推断方式一致,都是用分布函数作推断,称为分布推断。从分析传统的参数估计、假设检验特性来看,经典统计推断也可以视为分布推断。通常将置信上限看做置信度的函数。其反函数,即置信度是置信上界的函数,恰是分布函数,该分布恰是近年来引起许多学者兴趣的CD。在本文中,基于随机化估计(其分布是一CD)的概率密度函数,提出VDR检验。常见正态分布期望或方差的检验,多元正态分布期望的Hoteling检验等是其特例。VDR(vertical density representation)检验适合于多元分布参数检验,实现了非正态的多元线性变换分布族的参数检验。VDR构造的参数的置信域有最小Lebesgue测度。