北京交通流序列的重分形扩散熵分析

运用重分形扩散熵分析方法来分析北京交通拥堵指数的长程相关性和重分形特征.方法综合使用了扩散技术和Renyi熵来研究北京交通拥堵指数的标度行为.由于交通拥堵指数序列具有明显的周期性,故先选用傅里叶滤波去除序列的周期性,再进行重分形扩散熵分析.实验结果表明北京交通拥堵指数序列的极端波动显示出反相关性,同时拥堵指数序列具有较弱的重分形特征.     

基于分位点自回归模型的动态持续期风险估计

两笔交易之间的持续期可以用来度量资产的流动性,本文借鉴VaR的思想,提出了持续期风险DaR的定义,可以用来度量资产的流动性风险,并给出了两种DaR的静态估计方法。同时根据持续期数据的序列相关的特点,应用分位点自回归方法得到了DaR的一种有效的动态估计方法。最后对我国股票市场的单只股票的分笔交易数据进行了实证分析,结果表明由分位点自回归方法得出的DaR结果预测效果最好。     

中美两国公司债信用利差动态过程比较研究

选取中美两国2011年1月至2017年4月的公司债和国债月度交易数据,基于SV模型得到两国公司债的信用利差序列,进而对中美两国公司债的信用利差进行时间序列比较分析.实证发现,中国公司债信用利差序列表现出自回归和移动平均特征,而美国公司债信用利差序列则仅呈现自回归特征;在方差结构方面,中国公司债信用利差序列的残差不具有ARCH效应,而美国公司债信用利差序列的残差具有明显的ARCH效应.同时,对中美两国公司债信用利差建立VAR模型并进行脉冲响应分析,发现中美两国信用利差序列的相关性不强,对彼此的冲击的反应均较弱,为债券市场投资者构建跨国市场债券组合来分散信用风险提供决策支持.     

基于标度理论的股指时间序列相似性分析

股指时间序列的相似性分析是当前金融学研究的热点之一。为了提高股指时间序列相似性分析的准确度,从标度不变性、多重分形及波动聚集性三个层面定义了标度理论的度量指标,并基于此对股指序列进行表示。将分割后的每一序列子区间看作时间点,则分割、表示后的不同股指序列构成一个多指标的面板数据。基于面板数据特征及指标相对重要性,提出了一种新型的多指标面板数据相似性度量函数——复合距离函数,用以度量股指时间序列的相似性。聚类结果表明,相较于其他两种方法,基于标度理论和复合距离函数的相似性度量方法能够显著提高相似性度量的准确度,同时具有较强的稳健性。     

基于多分辨分析的沪深股市相关性分析

为了有效地揭示沪深股市不同交易周期股票交易的相关性,本文采用极大重叠离散小波变换,将上证指数和深圳成指高频收益率分解在不同的交易周期上,对各个周期的收益率采用semi-GPD模型作为其边缘分布分别进行拟合,在此基础上采用Copula函数方法建立同周期收益率的联合分布,度量了沪深两市同周期交易的相关性.实证表明,不同交易周期所表现出的相关性存在明显差异,并且随着交易期的增长,沪深两市非对称结构逐步明显。     

基于CEEMD与GA-SVR的猪肉价格集成预测模型

为提高猪肉价格预测的准确性,结合互补集合经验模态分解(CEEMD)的分解能力和基于遗传算法的支持向量回归(GA-SVR)的自适应预测功能,构建猪肉价格集成预测模型.首先为解决猪肉价格的复杂波动特征,通过CEEMD对猪肉价格分解得到本征模态函数(IMF)序列集;然后使用排序熵(PE)对IMF序列进行复杂度分析,进一步使用快速傅里叶变换方法(FFT)分解复杂度高的序列;再利用灰色关联度(GCD)对IMF序列集进行关联性分析,聚合相似IMF序列;最后基于各IMF序列的数据特征构建相应的GA-SVR预测模型,并将子序列的预测结果集成获得最终价格预测值.以中国集贸市场的猪肉价格为研究对象,实证结果表明,该集成预测模型在预测精度和方向性指标上,显著优于其他单预测模型和分解集成预测模型.     

股票收益率尾部相关性的Copula度量及模拟

股票收益率尾部相关性是研究金融市场关联性的重要内容.由于传统的τ、ρ等相关系数是对随机变量的全局度量,不适合用于收益率分布尾部这种局部特征的相关性度量.因此,在引入左尾(右尾)相关系数的基础上,讨论了它们的Copula度量及其相关性质.最后,通过计算机模拟分析了沪、深股指收益率尾部相关性的变化趋势,有效避免了Copula模型的设定困难,并得到了尾部相关性增强、相关不对称等结论.     

数据流的非平稳性度量

该文融合遍历论、粗粒化方法和信息论的观点研究数据流的非平稳性度量问题.引入了数据流的非平稳性度量的概念,给出了数据流非平稳性度量的有效的近似算法.数据流的非平稳性度量为0和1之间的实数,平稳性较好的数据流的非平稳性度量较小.作者将数据流的非平稳性度量应用到模型选择问题中,提出残差序列非平稳性度量最小化的模型选择标准.作者用数值试验检验了该文提出的数据流非平稳性度量的近似算法,并检验了其作为模型选择标准的能力.数值试验的结果表明,非平稳性度量是衡量数据流非平稳程度的一个合理指标,可以很好地区分趋势平稳数据和差分平稳数据,区分独立同分布序列、白噪声序列和鞅差序列.     

傅里叶级数引入环节的教学设计

本文在分析学情的基础上,对傅里叶级数引入环节进行教学设计,多角度展示了傅里叶级数的本质,意义和应用,为后续理论部分的讲解打好了基础.     

基于数据挖掘的保健药品关联性分析

信息技术的迅速发展以及数据库技术的日新月异,大量的信息成为当代的一个特点.保健药品企业存储了海量的交易数据,通常,面对这样的数据,人们往往无所适从,数据挖掘的方法向我们提供了对海量数据进行加工和处理的工具,基于企业交易海量数据,建立商品层面关联分析模型,并分别对订购商不加时间区分的进行商品关联分析、按月合并订购商订单后进行商品关联分析,此外,还进一步地进行了商品的序列关联分析.     

成分数据相关系数的计算方法

成分数据具有非常复杂的数学性质,很多传统的统计分析方法对其是失效的,因此,在研究中必须采用特殊处理和专门技术.着重讨论了成分数据相关系数的计算方法,由于普通数据的相关系数计算方法只适用于两组单变量数据,而传统的典型相关分析又鉴于成分数据的特殊性质而不能直接使用,故结合logratio变换和典型相关分析技术,提出了一种针对成分数据的相关系数计算方法,成功地解决了这一问题.     

The inverse scattering transform: Tools for the nonlinear fourier analysis and filtering of ocean surface waves

Surface wave data from the Adriatic Sea are analysed in the light of new data analysis techniques which may be viewed as a nonlinear generalization of the ordinary Fourier transform. Nonlinear Fourier analysis as applied herein arises from the exact spectral solution to large classes of nonlinear wave equations which are integrable by the inverse scattering transform (IST). Numerical methods are discussed which allow for implementation of the approach as a tool for the time series analysis of oceanic wave data. The case for unidirectional propagation in shallow water, where integrable nonlinear wave motion is governed by the Korteweg-deVries (KdV) equation with periodic/quasi-periodic boundary conditions, is considered. Numerical procedures given herein can be used to compute a nonlinear Fourier representation for a given measured time series. The nonlinear oscillation modes (the IST ‘basis functions’) of KdV obey a linear superposition law, just as do the sine waves of a linear Fourier series. However, the KdV basis functions themselves are highly nonlinear, undergo nonlinear interactions with each other and are distinctly non-sinusoidal. Numerical IST is used to analyse Adriatic Sea data and the concept of nonlinear filtering is applied to improve understanding of the dominant nonlinear interactions in the measured wavetrains.     

Measuring rank correlation coefficients between financial time series: A GARCH-copula based sequence alignment algorithm

This paper presents a novel four-stage algorithm for the measurement of the rank correlation coefficients between pairwise financial time series. In first stage returns of financial time series are fitted as skewed-t distributions by the generalized autoregressive conditional heteroscedasticity model. In the second stage, the joint probability density function (PDF) of the fitted skewed-t distributions is computed using the symmetrized Joe–Clayton copula. The joint PDF is then utilized as the scoring scheme for pairwise sequence alignment in the third stage. After solving the optimal sequence alignment problem using the dynamic programming method, we obtain the aligned pairs of the series. Finally, we compute the rank correlation coefficients of the aligned pairs in the fourth stage. To the best of our knowledge, the proposed algorithm is the first to use a sequence alignment technique to pair numerical financial time series directly, without initially transforming numerical values into symbols. Using practical financial data, the experiments illustrate the method and demonstrate the advantages of the proposed algorithm.     

On sharp estimates of the convergence of double Fourier–Bessel series

The problem of approximation of a differentiable function of two variables by partial sums of a double Fourier–Bessel series is considered. Sharp estimates of the rate of convergence of the double Fourier–Bessel series on the class of differentiable functions of two variables characterized by a generalized modulus of continuity are obtained. The proofs of four theorems on this issue, which can be directly applied to solving particular problems of mathematical physics, approximation theory, etc., are presented.     

The analysis of correlation integrals in terms of extremal value theory

In this paper we first give an overview of the methods of analysis of time series in terms of correlation integrals, which were developed for time series generated by deterministic systems. From the extremal value theory one obtains asymptotic information on the behaviour of the correlation integrals of time series generated by non-deterministic (mixing) systems. This leads to an analysis in terms of correlation integrals which is complementary to the estimation of dimension and entropy.Dedicated to the memory of Ricardo Mañé     

Characterizing the departure process of a single server queue from the embedded Markov renewal process at departures

In the literature, performance analyses of numerous single server queues are done by analyzing the embedded Markov renewal processes at departures. In this paper, we characterize the departure processes for a large class of such queueing systems. Results obtained include the Laplace–Stieltjes transform (LST) of the stationary distribution function of interdeparture times and recursive formula for {_cn ≡ the covariance between interdeparture times of lag n}. Departure processes of queues are difficult to characterize and for queues other than M/G/1 this is the first time that {_cn} can be computed through an explicit recursive formula. With this formula, we can calculate {_cn} very quickly, which provides deeper insight into the correlation structure of the departure process compared to the previous research. Numerical examples show that increasing server irregularity (i.e., the randomness of the service time distribution) destroys the short-range dependence of interdeparture times, while increasing system load strengthens both the short-range and the long-range dependence of interdeparture times. These findings show that the correlation structure of the departure process is greatly affected by server regularity and system load. Our results can also be applied to the performance analysis of a series of queues. We give an application to the performance analysis of a series of queues, and the results appear to be accurate. This revised version was published online in June 2006 with corrections to the Cover Date.     

Detection of low-dimensional chaos in buildings energy consumption time series

In this paper, nonlinear time series modeling techniques are applied to analyze building energy consumption data. The time series were obtained for the benchmark data set Proben 1, and comes from the first energy prediction contest, the Great Building Energy Predictor Shootout I, organized by ASHRAE. The phase space, which describes the evolution of the behavior of a nonlinear system, is reconstructed using the delay embedding theorem suggested by TAKENS. The embedding parameters, e.g. the delay time and the embedding dimension are estimated using the average mutual information (AMI) of the data and the false nearest neighbor (FNN) algorithm, respectively. Nonlinearity was detected, by applying the surrogate data sets method.Numerically estimated non-integral fractal dimension and a positive Lyapunov exponent are not necessarily sufficient indication of chaos; therefore we apply a more stringent criterion, developed by Gao and Zheng, which is based on the logarithmic displacement of time-dependent exponent curves, and show that these data are chaotic.Based on this analysis and proof, we then calculate the correlation dimension of the resulting attractor and the largest Lyapunov exponent. The correlation dimension 3.47 and largest Lyapunov exponent 0.047 are estimated. These results indicate that chaotic characteristics obviously exist in the specific energy consumption data set, and thus techniques based on phase space dynamics can be used to analyze and predict buildings energy use.     

基于组合赋权的灰色关联分析在水质评价中的应用研究

在灰色关联理论的基础上,提出了基于超标比与均方差组合赋权的水质综合评价方法,可有效的识别水体中的主要污染来源.通过改进数据标准化的方法,合理的减弱了水质监测数据中的奇点对评价结果的影响.将方法应用于白洋淀五个监测断面的水质综合评价,取得了合理的结果,表明方法可行.     

基于修正邓氏灰色关联度的设备费用影响因素分析

针对武器装备数据多为小样本情况,采用灰色关联度对其研制生产费用影响因素的确定,针对灰色关联度不能处理负相关和组合相关的问题,提出应用普通相关和逐步回归对灰色关联度结果进行修正,实例分析表明处理结果更符合实际情况.     

Non-linear properties of R–R distributions as a measure of heart rate variability

We analyze the dynamic quality of the R–R interbeat intervals of electrocardiographic signals from healthy people and from patients with premature ventricular contractions (PVCs) by applying different measure algorithms to standardised public domain data sets of heart rate variability. Our aim is to assess the utility of these algorithms for the above mentioned purposes.

Long and short time series, 24 and 0.50 h respectively, of interbeat intervals of healthy and PVC subjects were compared with the aim of developing a fast method to investigate their temporal organization.

Two different methods were used: power spectral analysis and the integral correlation method.

Power spectral analysis has proven to be a powerful tool for detecting long-range correlations. If it is applied in a short time series, power spectra of healthy and PVC subjects show a similar behavior, which disqualifies power spectral analysis as a fast method to distinguish healthy from PVC subjects.

The integral correlation method allows us to study the fractal properties of interbeat intervals of electrocardiographic signals.

The cardiac activity of healthy and PVC people stems from dynamics of chaotic nature characterized by correlation dimensions d_f equal to 3.40±0.50 and 5.00±0.80 for healthy and PVC subjects respectively.

The methodology presented in this article bridges the gap between theoretical and experimental studies of non-linear phenomena. From our results we conclude that the minimum number of coupled differential equations to describe cardiac activity must be six and seven for healthy and PVC individuals respectively.

From the present analysis we conclude that the correlation integral method is particularly suitable, in comparison with the power spectral analysis, for the early detection of arrhythmias on short time (0.5 h) series.