基于修剪平均的神经网络集成时序预测方法

由于复杂时序存在结构性断点和异常值等问题，往往导致预测模型训练效果不佳，并可能出现极端预测值的情况。为此，本文提出了基于修剪平均的神经网络集成预测方法。该方法首先从训练数据中生成多组训练集，然后分别训练多个神经网络预测模型，最后将多个神经网络的预测结果使用修剪平均策略进行集成。相较于简单平均策略而言，修剪平均策略不容易受到极值的影响，能够使集成模型获得鲁棒性强的预测效果。在实证研究中，本文构造了两种神经网络集成预测模型，分别为基于修剪平均的自举神经网络集成模型(Trimmed Average based Bootstrap Neural Network Ensemble, TA-BNNE)和基于修剪平均的蒙特卡洛神经网络集成模型(Trimmed Average based Monte Carlo Neural Network Ensemble, TA-MCNNE)，并采用这两种模型对NN3竞赛数据集进行预测，结果表明在常规和复杂数据集上，修剪平均策略比简单平均策略具有更好的预测精度。此外，本文将所提出的集成模型与NN3的前十名模型进行比较，发现两种模型在全部数据集上均超过了第6名，在复杂数据集上的表现均超过了第1名，进一步验证本文所提方法的有效性。     

基于小波分析的石油价格长期趋势预测方法及其实证研究

本文将小波方法引入到油价长期趋势的预测中,利用小波多尺度分析的功能,提出了一种可以较为准确地根据油价时序列预测其未来长期走势的方法。这种方法的优点在于可以准确地提取油价的长期趋势,从总体上把握油价的非线性波动特征,从而能够很好地利用油价时间序列的历史数据,开展对未来一段时期内的多步预测。实证研究中,对Brent油价开展了时间跨度为1年的趋势预测,并将预测结果与ARIMA、GARCH、Holtwinters等方法得到的结果进行了比较,表明了基于小波分析的长期趋势预测法的预测能力是其他方法所不能比拟的,反映了本文所建立的石油价格长期趋势预测方法的有效性。     

基于近邻互信息的SVM-GARCH股票价格预测模型研究

为了克服传统线性模型分析处理收益率数据非线性因素的不足,本文提出一种新的基于近邻互信息特征选择的SVM-GARCH预测模型。该模型利用SVM处理高维非线性数据的优势,不仅包含了股指序列自身的历史数据信息,而且通过近邻互信息的方式融合了与目标股指数据关系密切的周边证券市场的相关变化信息。仿真实验结果表明,该模型在时序数据除噪、趋势判别以及预测的精确度等方面均优于传统的ARMA-GARCH模型。     

FIRM DEFAULT AND AGGREGATE FLUCTUATIONS

This paper studies the relationship between macroeconomic fluctuations and corporate defaults while conditioning on industry affiliation and an extensive set of firm‐specific factors. By using a panel data set for virtually all incorporated Swedish businesses over 1990–2009, a period which includes a full‐scale banking crisis, we find strong evidence for a substantial and stable impact from aggregate fluctuations on business defaults. A standard logit model with financial ratios augmented with macroeconomic factors can account surprisingly well for the outburst in business defaults during the banking crisis, as well as the subsequent fluctuations in default frequencies. Moreover, the effects of macroeconomic variables differ across industries in an economically intuitive way. Out‐of‐sample evaluations show that our approach is superior to models that exclude macro information and standard well‐fitting time‐series models. Our analysis shows that firm‐specific factors are useful in ranking firms’ relative riskiness, but that macroeconomic factors are necessary to understand fluctuations in the absolute risk level.     

Developing inflation scenarios

The purpose of this paper is to describe a scenario-based approach of forecasting future levels of inflation as it has been developed and applied in the context of retirement income planning. The circumstances and settings for the use of time series analysis in the generation of scenarios are discussed.     

Asymptotic Distribution Theory for Nonparametric Entropy Measures of Serial Dependence

Entropy is a classical statistical concept with appealing properties. Establishing asymptotic distribution theory for smoothed nonparametric entropy measures of dependence has so far proved challenging. In this paper, we develop an asymptotic theory for a class of kernel‐based smoothed nonparametric entropy measures of serial dependence in a time‐series context. We use this theory to derive the limiting distribution of Granger and Lin's (1994) normalized entropy measure of serial dependence, which was previously not available in the literature. We also apply our theory to construct a new entropy‐based test for serial dependence, providing an alternative to Robinson's (1991) approach. To obtain accurate inferences, we propose and justify a consistent smoothed bootstrap procedure. The naive bootstrap is not consistent for our test. Our test is useful in, for example, testing the random walk hypothesis, evaluating density forecasts, and identifying important lags of a time series. It is asymptotically locally more powerful than Robinson's (1991) test, as is confirmed in our simulation. An application to the daily S&P 500 stock price index illustrates our approach.     

动静态属性数据相结合的客户分类方法研究

研究了一种客户动态、静态属性数据相结合的客户分类方法。提出了客户时间序列的加权处理方法,并应用客户时间序列的统计特征作为聚类特征向量,采用混合式遗传算法对客户聚类,使每一类客户具有相似的时序特征。在此基础上将聚类结果与客户的静态属性数据相结合,对客户进一步分类。实验结果表明,与传统的基于静态属性数据的客户分类方法相比,本文的方法提高了客户分类的准确性。     

ICT-enabled time performance: an investigation of value creation mechanisms

The study presented in this paper aimed at checking whether companies that embed information and communication technologies (ICT)-enabled time performance into their product offering can achieve better economic outcomes from technology adoption. Indeed, it is still questionable whether technology adoption results in a superior profitability, especially if such an improvement is achieved through the use of functional ICT applications. In this study, we assume that a better alignment among ICT investments, improvements of the logistics process and the value proposition of the firm can lead towards a superior economic performance. A survey was conducted and 180 usable questionnaires were collected from companies engaged in the electronics and vehicle manufacturing industries. Data were analysed through the structural equation modeling approach. The results show that improvements in speed and dependability, achieved through technology adoption, can lead to a better economic performance if they are embedded into superior after-sales services and/or into an improved product offering.     

Empirical Limits for Time Series Econometric Models

This paper characterizes empirically achievable limits for time series econometric modeling and forecasting. The approach involves the concept of minimal information loss in time series regression and the paper shows how to derive bounds that delimit the proximity of empirical measures to the true probability measure (the DGP) in models that are of econometric interest. The approach utilizes joint probability measures over the combined space of parameters and observables and the results apply for models with stationary, integrated, and cointegrated data. A theorem due to Rissanen is extended so that it applies directly to probabilities about the relative likelihood (rather than averages), a new way of proving results of the Rissanen type is demonstrated, and the Rissanen theory is extended to nonstationary time series with unit roots, near unit roots, and cointegration of unknown order. The corresponding bound for the minimal information loss in empirical work is shown not to be a constant, in general, but to be proportional to the logarithm of the determinant of the (possibility stochastic) Fisher–information matrix. In fact, the bound that determines proximity to the DGP is generally path dependent, and it depends specifically on the type as well as the number of regressors. For practical purposes, the proximity bound has the asymptotic form (K/2)log n, where K is a new dimensionality factor that depends on the nature of the data as well as the number of parameters in the model. When ‘good’ model selection principles are employed in modeling time series data, we are able to show that our proximity bound quantifies empirical limits even in situations where the models may be incorrectly specified. One of the main implications of the new result is that time trends are more costly than stochastic trends, which are more costly in turn than stationary regressors in achieving proximity to the true density. Thus, in a very real sense and quantifiable manner, the DGP is more elusive when there is nonstationarity in the data. The implications for prediction are explored and a second proximity theorem is given, which provides a bound that measures how close feasible predictors can come to the optimal predictor. Again, the bound has the asymptotic form (K/2)log n, showing that forecasting trends is fundamentally more difficult than forecasting stationary time series, even when the correct form of the model for the trends is known.     

基于幂律型分布的动态VaR模型及实证研究

针对金融资产回报时间序列的尖峰厚尾性和波动集聚性,提出了基于AR(1)-GARCH(1,1)模型与幂律型分布相结合计算VaR的方法。用GARCH模型对时间序列建模刻画波动集聚性,用基于幂律型分布的扩展形式拟合GARCH模型的残差分布尾部,刻画回报时间序列的厚尾特征,二者结合更好地描述回报时序的动态波动现象。对上证综指进行实证分析,结果表明本文提出的方法比基于正态分布的GARCH模型和静态幂律尾法更精确。     

Flood Hazard and Flood Risk Assessment Using a Time Series of Satellite Images: A Case Study in Namibia

In this article, the use of time series of satellite imagery to flood hazard mapping and flood risk assessment is presented. Flooded areas are extracted from satellite images for the flood‐prone territory, and a maximum flood extent image for each flood event is produced. These maps are further fused to determine relative frequency of inundation (RFI). The study shows that RFI values and relative water depth exhibit the same probabilistic distribution, which is confirmed by Kolmogorov‐Smirnov test. The produced RFI map can be used as a flood hazard map, especially in cases when flood modeling is complicated by lack of available data and high uncertainties. The derived RFI map is further used for flood risk assessment. Efficiency of the presented approach is demonstrated for the Katima Mulilo region (Namibia). A time series of Landsat‐5/7 satellite images acquired from 1989 to 2012 is processed to derive RFI map using the presented approach. The following direct damage categories are considered in the study for flood risk assessment: dwelling units, roads, health facilities, and schools. The produced flood risk map shows that the risk is distributed uniformly all over the region. The cities and villages with the highest risk are identified. The proposed approach has minimum data requirements, and RFI maps can be generated rapidly to assist rescuers and decisionmakers in case of emergencies. On the other hand, limitations include: strong dependence on the available data sets, and limitations in simulations with extrapolated water depth values.     

基于GJR-GARCH的VaR模型及其在上海证券市场的实证研究

针对风险价值VaR的一般参数方法都是对称的,其在处理非对称时间序列时存在着局限性,本文提出了非对称的VaR计算模型,并以上海证券市场为对象进行了实证研究,结果表明基于非对称的VaR计算模型优于对称的VaR计算模型。     

能源价格对制造业能源强度调节效应的实证研究

能耗过高影响中国制造业的可持续发展,能源价格是市场体制中调节能源强度的重要因素和手段.引入价格作为调节变量,研究其对于技术进步和能源消费结构对行业能源强度影响的调节效应.运用1995年～2005年的数据,采用层次回归法对制造业24个重要行业的能源价格与能源强度的关系进行实证研究.结果表明,对大多数行业,能源价格的提升并未明显降低能源强度,能源价格对于技术进步对能源强度影响的调节效应不显著,能源价格的调节效应更多地表现为促进能源消费结构转化来降低行业能源强度.     

基于EMD与DVG的非线性时间序列预测模型及其应用研究

针对具有非线性和不稳定性的时间序列,提出一种结合经验模态分解(EMD)、有向可见图(DVG)网络的动态预测模型。利用经验模态分解将原时间序列分解为多个固有模态函数(IMF),然后对分解后的高频和低频IMF利用快速傅里叶变换得到各自的周期;依据每个周期,从原时间序列的尾部截取长短不一的子序列,然后采用有向可见图算法转换为多个有向网络,利用随机游走在每个有向网络中寻找与时间序列最后一个节点相似的节点;最后,依据平行线法,预测时间序列的下一个数值。原油价格的时间序列是一类典型的具有非线性和不稳定性的序列,利用此模型对WTI原油每日价格进行实证分析。研究结果表明,此模型不但可以有效地预测时间序列的变化趋势,而且具有较高的预测精度。     

The competitive advantage and strategic configuration of knowledge-intensive, small- and medium-sized multinationals: a modified resource-based view

This study employs a modified resource-based approach to examine the competitive advantage enjoyed by knowledge-intensive, small- and medium-sized multinationals (KI-SMMs). While the resource-based view addresses superior capabilities only, this paper examines both superior and inferior capabilities and their resulting sustainable competitive advantage.Compared to larger knowledge-intensive multinationals, KI-SMMs possess more inferior than superior core capabilities. Despite this handicap, the paper demonstrates how KI-SMMs compete globally by leveraging their relatively superior R&D capabilities and by choosing a strategic configuration that allows them to compete internationally despite their relatively inferior capabilities in marketing and production activities.Our results show that KI-SMMs internalize R&D activities, which are their core capabilities, externalize production activities, in this case noncore capabilities, and internalize marketing activities, for which they have an inferior capacity, but which are, arguably, core capabilities. KI-SMMs compensate for their inferior capabilities in marketing activities through the use of a unique business model which focuses repeat sales to customers with whom a low number of high-value transactions can be maintained.     

Bayesian Temporal Source Attribution of Foodborne Zoonoses: Campylobacter in Finland and Norway

Statistical source attribution approaches of food‐related zoonoses can generally be based on reported diagnosed human cases and surveillance results from different food sources or reservoirs of bacteria. The attribution model, or probabilistic classifier, can thus be based on the (sub)typing information enabling comparison between human infections and samples derived from source surveillance. Having time series of both data allows analyzing temporal patterns over time providing a repeated natural experiment. A Bayesian approach combining both sources of information over a long time series is presented in the case of Campylobacter in Finland and Norway. The full model is transparently presented and derived from the Bayes theorem. Previous statistical source attribution approaches are here advanced (1) by explicit modeling of the cases not associated with any of the sources under surveillance over time, (2) by modeling uncertain prevalence in a food source by bacteria type over time, and (3) by implementing formal model fit assessment using posterior predictive discrepancy functions. Large proportion of all campylobacteriosis can be attributed to broiler, but considerable uncertainty remains over time. The source attribution is inherently incomplete if only the sources under surveillance are included in the model. All statistical source attribution approaches should include a model fit assessment for judgment of model performance with respect to relevant quantities of interest. It is especially relevant when the model aims at a synthesis of several incomplete information sources under significant uncertainty of explanatory variables.     

投资组合优化中的泰勒级数收敛性与展开点选择问题研究

研究了泰勒级数对效用函数的收敛条件,力求能够使得泰勒级数成为效用函数的合理近似,从而保证投资组合优化问题的近似解收敛于真实解,最终实现最大化期望效用的投资组合优化问题得以有效解决。在期望效用最大化的泰勒级数近似模型基础上,以HARA效用函数为背景,得到了收益率相对泰勒级数展开点的偏离程度决定了泰勒级数收敛性质的结论,进而提出了合理选择泰勒级数展开点以保证收敛性的方法,该方法意味着在收益率分布具有正偏度的情况下,以往通行的在收益率数学期望处展开泰勒级数的方法不具有合理性,上述分析结论通过数据分析得到了验证。     

Analysis of a batch machine fed by an unreliable discrete machine

This paper analyses the behaviour of a manufacturing line consisting of two machines in series where the first machine processes lots one at a time, and is subject to failure and takes a random amount of time to repair when it fails, and the second machine is a perfectly reliable batch machine. A control limit policy is adopted to determine lot sizes for the batch machine. When the batch machine completes processing, if the number of lots in the buffer is greater than or equal to the critical number (Q?), all the lots in the buffer are loaded immediately, otherwise the batch machine waits until Q lots are accumulated. An embedded discrete time Markov-chain approach is proposed, and recursive approaches are developed to derive necessary performance measures. A numerical example explains how to obtain the optimal value of a critical number minimizing the cost functions.     

Strategy,core competencies and human resource development

HRD's traditional approach to competency focuses on superior individual performance. The concept of a core competency in the strategic literature, as reviewed here, considers it as an organizational capability that provides sustained competitive advantage. That model is examined and implications for HRD practice noted.