On weak convergence of long-range-dependent traffic processes

A possible model for communication traffic is that the amount of work arriving in successive time intervals is jointly Gaussian. This model seems to fly in the face of certain obvious and characteristic features of real traffic, such as the fact that it arrives in discrete bundles and that there is often a non-zero probability of zero traffic in a time interval of significant length. Also, the Gaussian model allows the possibility of negative traffic, which is clearly unrealistic. As the number of sources of traffic increases and the quantity of traffic in communication networks increases, however, under suitable conditions, the deviation between the distribution of real traffic and the Gaussian model will become less. The appropriate concept of topology/convergence must be used or the result will be meaningless. To identify an appropriate convergence framework, the performance statistics associated with a network, namely cell loss, delay, and, in general, statistics which can be expressed in terms of the network buffers which accumulate in the network may be used as a guide. Weak convergence of probability measures has the property that when the probability measures of traffic processes converge to that of a certain traffic process, the distribution of their performance characteristics, such as buffer occupancy, also converges in the same sense to the performance of the system to which they were converging. Real traffic appears, unambiguously, to be long-range dependent. There is an interesting example where aggregation of traffic does not seem to produce convergence to the queueing behaviour expected of Gaussian traffic, at any rate the tail characteristics do not converge to those of the Gaussian result. However, in Section 4, it is shown that if the variance of one traffic stream is finite and as a proportion of the variance of the whole traffic volume tends to zero, then the traffic in networks can be expected to converge to Gaussian in the sense of weak convergence of probability measures. It is then shown that, as a consequence, the traffic in the paradoxical example does converge in this sense also. The paradox is explained by noticing that asymptotic tail behaviour may become increasingly irrelevant as traffic is aggregated. This fact should sound a warning concerning the cavalier use of tail-behaviour as an indication of performance. Long-range dependence apparently places no inhibition on convergence to Gaussian behaviour. Convergence to a Gaussian distribution of increasing aggregates of traffic is only shown to occur for discrete time models. In fact it appears that continuous time Gaussian models do not share this property and their use for modelling real traffic may be problematic.     

分位数回归与上证综指VaR研究

把极端分位数所具有的行为特征应用到VaR的研究中，建立上海股市收益率的条件分位数回归模型，描述其在极端分位数下的变化趋势。同时选取适当的尾部模型，并在此基础之上应用外推法预测非常极端分位数下的条件VaR，并与直接由分位数回归模型预测的结果进行比较。结果表明：两种方法得到的结果变化趋势都是一致的，由外推法预测的结果相对小一些。     

Lévy‐based Modelling in Brain Imaging

Abstract. A substantive problem in neuroscience is the lack of valid statistical methods for non‐Gaussian random fields. In the present study, we develop a flexible, yet tractable model for a random field based on kernel smoothing of a so‐called Lévy basis. The resulting field may be Gaussian, but there are many other possibilities, including random fields based on Gamma, inverse Gaussian and normal inverse Gaussian (NIG) Lévy bases. It is easy to estimate the parameters of the model and accordingly to assess by simulation the quantiles of test statistics commonly used in neuroscience. We give a concrete example of magnetic resonance imaging scans that are non‐Gaussian. For these data, simulations under the fitted models show that traditional methods based on Gaussian random field theory may leave small, but significant changes in signal level undetected, while these changes are detectable under a non‐Gaussian Lévy model.     

Reliability evaluation subject to assured accuracy rate and time for stochastic unreliable-node computer networks

In many real-life networks such as computer networks, branches and nodes have multi-state capacity, lead time, and accuracy rate. The network with unreliable nodes is more complex to evaluate the reliability because node failure results in the disabled of adjacent branches. Such a network is named a stochastic unreliable-node computer network (SUNCN). Under the strict assumption that each component (branch and node) has a deterministic capacity, the quickest path (QP) problem is to find a path sending a specific amount of data with minimum transmission time. The accuracy rate is a critical index to measure the performance of a computer network because some packets are damaged or lost due to voltage instability, magnetic field effects, lightning, etc. Subject to both assured accuracy rate and time constraints, this paper extends the QP problem to discuss the system reliability of an SUNCN. An efficient algorithm based on a graphic technique is proposed to find the minimal capacity vector meeting such constraints. System reliability, the probability to send a specific amount of data through multiple minimal paths subject to both assured accuracy rate and time constraints, can subsequently be computed.     

Pitfalls in using Weibull tailed distributions

By assuming that the underlying distribution belongs to the domain of attraction of an extreme value distribution, one can extrapolate the data to a far tail region so that a rare event can be predicted. However, when the distribution is in the domain of attraction of a Gumbel distribution, the extrapolation is quite limited generally in comparison with a heavy tailed distribution. In view of this drawback, a Weibull tailed distribution has been studied recently. Some methods for choosing the sample fraction in estimating the Weibull tail coefficient and some bias reduction estimators have been proposed in the literature. In this paper, we show that the theoretical optimal sample fraction does not exist and a bias reduction estimator does not always produce a smaller mean squared error than a biased estimator. These are different from using a heavy tailed distribution. Further we propose a refined class of Weibull tailed distributions which are more useful in estimating high quantiles and extreme tail probabilities.     

Extreme quantiles and tail index of a distribution based on kernel estimator

In this paper, we reveal the relationship between the tail exponent introduced by Parzen (1979) and tail index for a distribution function. Furthermore, we analyze the domain of attraction of the weighted sum of the distributions and its tail index. We show that the extreme quantiles can estimate directly, through knowing only the tail index of the kernel distribution function used in estimating the distribution function. Moreover, we give a smoothing parameter of extreme quantiles, which does not depend on any distribution function. The simulations and the application to reals data show that the proposed smoothed parameter gives better results for a heavy-tailed distribution, and for small sizes sample in extremes level.     

Copula‐based semiparametric analysis for time series data with detection limits

The analysis of time series data with detection limits is challenging due to the high‐dimensional integral involved in the likelihood. Existing methods are either computationally demanding or rely on restrictive parametric distributional assumptions. We propose a semiparametric approach, where the temporal dependence is captured by parametric copula, while the marginal distribution is estimated non‐parametrically. Utilizing the properties of copulas, we develop a new copula‐based sequential sampling algorithm, which provides a convenient way to calculate the censored likelihood. Even without full parametric distributional assumptions, the proposed method still allows us to efficiently compute the conditional quantiles of the censored response at a future time point, and thus construct both point and interval predictions. We establish the asymptotic properties of the proposed pseudo maximum likelihood estimator, and demonstrate through simulation and the analysis of a water quality data that the proposed method is more flexible and leads to more accurate predictions than Gaussian‐based methods for non‐normal data. The Canadian Journal of Statistics 47: 438–454; 2019 © 2019 Statistical Society of Canada     

基于高斯谱聚类的风险商户聚类分析

随着电子支付的普及,市场涌现出越来越多的第三方支付平台,而当前关于第三方支付平台商户风险方面的研究相对较少。故本文提出基于高斯谱聚类的风险商户聚类方法,首先使用高斯混合模型构建交易-交易群体的双模网络;其次借助网络中信息传递的思想构建“商户-交易群体网络”的双模网络;再次使用双模网络聚类方法中的谱聚类方法同时对网络中的两类节点聚类,对商户节点聚类的结果可区分出不同风险级别的商户,对交易群体节点聚类的结果可以进一步描述风险商户的交易特征;最后本文分别在模拟数据和某第方支付平台的实际数据中验证了模型的有效性。实验结果表明,本文提出的方法不仅可以准确地区分出不同风险级别的商户群体,而且能总结归纳风险商户的交易特征,为风险商户的监管提供参考。     

Tail index and second-order parameters’ semi-parametric estimation based on the log-excesses

In this paper we are interested in the derivation of the asymptotic and finite-sample distributional properties of a ‘quasi-maximum likelihood’ estimator of a ‘scale’ second-order parameter β, directly based on the log-excesses of an available sample. Such estimation is of primordial importance for the adaptive selection of the optimal sample fraction to be used in the classical semi-parametric tail index estimation as well as for the reduced-bias estimation of the tail index, high quantiles and other parameters of extreme or even rare events. An application in the area of survival analysis is provided, on the basis of a data set on males diagnosed with cancer of the tongue.     

Distribution-free confidence intervals for quantiles in small samples

Estimators for quantiles based on linear combinations of order statistics have been proposed by Harrell and Davis(1982) and kaigh and Lachenbruch (1982). Both estimators have been demonstrated to be at least as efficient for small sample point estimation as an ordinary sample quantile estimator based on one or two order statistics: Distribution-free confidence intervals for quantiles can be constructed using either of the two approaches. By means of a simulation study, these confidence intervals have been compared with several other methods of constructing confidence intervals for quantiles in small samples. For the median, the Kaigh and Lachenbruch method performed fairly well. For other quantiles, no method performed better than the method which uses pairs of order statistics.     

Using Multinomial Mixture Models to Cluster Internet Traffic

The paper considers the clustering of two large sets of Internet traffic data consisting of information measured from headers of transmission control protocol packets collected on a busy arc of a university network connecting with the Internet. Packets are grouped into 'flows' thought to correspond to particular movements of information between one computer and another. The clustering is based on representing the flows as each sampled from one of a finite number of multinomial distributions and seeks to identify clusters of flows containing similar packet‐length distributions. The clustering uses the EM algorithm, and the data‐analytic and computational details are given.     

Semiparametric estimation of regression quantiles with application to standardizing weight for height and age in US children

The appropriate interpretation of measurements often requires standardization for concomitant factors. For example, standardization of weight for both height and age is important in obesity research and in failure-to-thrive research in children. Regression quantiles from a reference population afford one intuitive and popular approach to standardization. Current methods for the estimation of regression quantiles can be classified as nonparametric with respect to distributional assumptions or as fully parametric. We propose a semiparametric method where we model the mean and variance as flexible regression spline functions and allow the unspecified distribution to vary smoothly as a function of covariates. Similarly to Cole and Green, our approach provides separate estimates and summaries for location, scale and distribution. However, similarly to Koenker and Bassett, we do not assume any parametric form for the distribution. Estimation for either cross-sectional or longitudinal samples is obtained by using estimating equations for the location and scale functions and through local kernel smoothing of the empirical distribution function for standardized residuals. Using this technique with data on weight, height and age for females under 3 years of age, we find that there is a close relationship between quantiles of weight for height and age and quantiles of body mass index (BMI=weight/height²) for age in this cohort.     

Control variates and importance sampling for efficient bootstrap simulations

Importance sampling and control variates have been used as variance reduction techniques for estimating bootstrap tail quantiles and moments, respectively. We adapt each method to apply to both quantiles and moments, and combine the methods to obtain variance reductions by factors from 4 to 30 in simulation examples.We use two innovations in control variates—interpreting control variates as a re-weighting method, and the implementation of control variates using the saddlepoint; the combination requires only the linear saddlepoint but applies to general statistics, and produces estimates with accuracy of order n ^-1/2 B ^-1, where n is the sample size and B is the bootstrap sample size.We discuss two modifications to classical importance sampling—a weighted average estimate and a mixture design distribution. These modifications make importance sampling robust and allow moments to be estimated from the same bootstrap simulation used to estimate quantiles.     

Extreme quantile estimation based on financial time series

Estimation of market risk is an important problem in finance. Two well-known risk measures, viz., value at risk and median shortfall, turn out to be extreme quantiles of the marginal distribution of asset return. Time series on asset returns are known to exhibit certain stylized facts, such as heavy tails, skewness, volatility clustering, etc. Therefore, estimation of extreme quantiles in the presence of such features in the data seems to be of natural interest. It is difficult to capture most of these stylized facts using one specific time series model. This motivates nonparametric and extreme value theory-based estimation of extreme quantiles that do not require exact specification of the asset return model. We review these quantile estimators and compare their known properties. Their finite sample performance are compared using Monte Carlo simulation. We propose a new estimator that exhibits encouraging finite sample performance while estimating extreme quantile in the right tail region.     

Semi-parametric expected shortfall forecasting in financial markets

Bayesian methods have proved effective for quantile estimation, including for financial Value-at-Risk forecasting. Expected shortfall (ES) is a competing tail risk measure, favoured by the Basel Committee, that can be semi-parametrically estimated via asymmetric least squares. An asymmetric Gaussian density is proposed, allowing a likelihood to be developed, that facilitates both pseudo-maximum likelihood and Bayesian semi-parametric estimation, and leads to forecasts of quantiles, expectiles and ES. Further, the conditional autoregressive expectile class of model is generalised to two fully nonlinear families. Adaptive Markov chain Monte Carlo sampling schemes are developed for the Bayesian estimation. The proposed models are favoured in an empirical study forecasting eight financial return series: evidence of more accurate ES forecasting, compared to a range of competing methods, is found, while Bayesian estimated models tend to be more accurate. However, during a financial crisis period most models perform badly, while two existing models perform best.     

Modeling service-time distributions for queueing network simulation

Shapes of service-time distributions in queueing network models have a great impact on the distribution of system response-times. It is essential for the analysis of response-time distribution that the modeled service-time distributions have the correct shape. Tradionally modeling of service-time distributions is based on a parametric approach by assuming a specific distribution and estimating its parameters. We introduce an alternative approach based on the principles of exploratory data analysis and nonparametric data modeling. The proposed method applies nonlinear data transformation and resistant curve fitting. The method can be used in cases, where the available data is a complete sample, a histogram, or the mean and a set of 5-10 quantiles. The reported results indicate that the proposed method is able to approximate the distribution of measured service times so that accurate estimates for quantiles of the response-time distribution are obtained     

A method for estimating parameters and quantiles of the three-parameter inverse Gaussian distribution based on statistics invariant to unknown location

The inverse Gaussian (IG) distribution, also known as the Wald distribution, is a long-tailed positively skewed distribution and a well-known lifetime distribution. In this paper, we propose an efficient method of estimation for the parameters and quantiles of the three-parameter IG distribution, which is based on statistics invariant to unknown location. Through a Monte Carlo simulation study, we then show that the proposed method performs well compared with other prominent methods in terms of bias and variance. Finally, we present two illustrative examples.     

Comparing two dependent groups via quantiles

This paper considers two general ways dependent groups might be compared based on quantiles. The first compares the quantiles of the marginal distributions. The second focuses on the lower and upper quantiles of the usual difference scores. Methods for comparing quantiles have been derived that typically assume that sampling is from a continuous distribution. There are exceptions, but generally, when sampling from a discrete distribution where tied values are likely, extant methods can perform poorly, even with a large sample size. One reason is that extant methods for estimating the standard error can perform poorly. Another is that quantile estimators based on a single-order statistic, or a weighted average of two-order statistics, are not necessarily asymptotically normal. Our main result is that when using the Harrell–Davis estimator, good control over the Type I error probability can be achieved in simulations via a standard percentile bootstrap method, even when there are tied values, provided the sample sizes are not too small. In addition, the two methods considered here can have substantially higher power than alternative procedures. Using real data, we illustrate how quantile comparisons can be used to gain a deeper understanding of how groups differ.     

A Multivariate Quantile Predictor

ABSTRACT

We introduce a nonparametric quantile predictor for multivariate time series via generalizing the well-known univariate conditional quantile into a multivariate setting for dependent data. Applying the multivariate predictor to predicting tail conditional quantiles from foreign exchange daily returns, it is observed that the accuracy of extreme tail quantile predictions can be greatly improved by incorporating interdependence between the returns in a bivariate framework. As a special application of the multivariate quantile predictor, we also introduce a so-called joint-horizon quantile predictor that is used to produce multi-step quantile predictions in one-go from univariate time series realizations. A simulation example is discussed to illustrate the relevance of the joint-horizon approach.