优化离散灰色幂模型及其应用

考虑已有的灰色预测模型主要能对指数型发展系统或幂函数型发展系统进行模拟预测,本文构建了一种不仅能够模拟指数型和幂函数型的发展系统,并且能够体现出二者之间的相互作用关系的离散灰色幂模型;并针对初始条件对离散灰色幂模型模拟精度的影响,首先给出了离散灰色幂模型的建模步骤,然后以平均相对误差最小化为目标、参数之间的关系为约束条件,构建了离散灰色幂模型初始条件的优化模型,实现对离散灰色幂模型初始条件的优化。结果表明,优化的离散灰色幂模型使得平均相对误差在理论上达到了最小化,其模拟精度和预测精度都高于传统模型。最后,通过中国网络购物人数数据预测和仿真数据分析,说明了本文优化方法的有效性和适用性。     

Aligning supply chain collaboration using Analytic Hierarchy Process

The significance of collaboration among supply chain members has been sufficiently stressed in the recent literature as a powerful tool for increasing accuracy of demand forecasts and for consequent cost reductions. Since it has been recognized that naïve forecasting is no longer cost efficient, Supply Chain (SC) members have found it very important to exchange relevant information that will help improve accuracy of demand forecasting. This information differs widely in terms of their characteristics. For example, some information (e.g. historic sales data) that is cheap to exchange may not contribute to a great increase in forecast accuracy. Similarly, some information may not be very reliable (e.g. demand forecast by individual SC members). In general, there is a trade-off in the kind of information required and the kind of information exchanged. This study analyses these trade-offs using an Analytic Hierarchy Process (AHP) model. The model is then implemented based on case studies conducted in two manufacturing firms. The AHP model ranks available information in terms of their contributions to improve forecast accuracy, and can provide vital clues to SC partners for preparing exchangeable data. From the case studies using AHP model, it was proved that using the preferred SC data, the firms could enhance forecasts accuracy. This in turn can help the firms to make decisions on SC collaborative arrangements for information exchange.     

考虑促销因素的医药电商平台需求预测研究

医药电商平台需求预测涉及到药品自身属性及电商平台推出的各种促销活动,本文针对以上影响药品销量的因素提出了时间序列-机器学习组合模型对医药电商平台进行需求预测。传统研究促销因素的需求预测文献将促销阶段商品销量拆分为常规销量和促销增量的线性组合,本文首先拟合各药品促销阶段的常规销量,根据各药品常规销量时间序列数据及服用周期,使用SARIMA模型拟合药品的常规销量预测值,并将常规销量预测值与商品促销特征数据一同输入XGBoost模型进行集成学习预测。本文使用国内某医药电商平台真实销售数据测试组合模型的有效性,结果显示组合预测模型的预测效果相比其他三种传统预测模型更优。此外,本文验证了不同折扣力度下组合预测模型的有效性,以及促销变量在预测模型中的有效性,同时研究了数据共享策略在需求预测中的应用场景,结果显示预测模型在引入促销变量和采用数据共享策略后都能显著降低模型的预测误差。     

AN ADAPTIVE MULTIVARIATE APPROACH TO TIME SERIES FORECASTING

In recent years, time series analysts have shifted their interest from univariate to multivariate forecasting approaches. Among them, the Box-Jenkins transfer function process and the state space method have received the most attention. This paper presents a simplified approach that embodies some desirable features of existing methods. It stresses empirical analysis, has a unified modeling structure, is easily applicable, and is adaptive to changes without necessitating prior information on the evolution of a system under study. The core of the method relies on the Carbone-Longini adaptive estimation procedure (AEP). Results of a comparative study based on the well-known Lydia E. Pinkham data and the Box-Jenkins sales/leading indicator data illustrate the merits of multivariate AEP in improving forecasting accuracy while simplifying the analysis process. Subject Area: Forecasting.     

基于网络搜索信息的农村水环境质量灰色预测模型

针对农村水环境直接监测数据相对缺乏、间接数据难以有效引入的问题,综合灰色关联分析和预测模型,提出一种基于网络搜索信息的农村水环境质量灰色预测模型。首先,综合考虑数据重要性和可获得性,确定农村水环境相关的搜索关键词清单;然后,采用主成分分析法提取搜索关键词的主要特征,构建初始网络搜索变量,并利用灰色绝对关联度衡量各初始网络搜索变量与水环境质量之间的关联程度。在此基础上,构建不同频率数据的多变量离散灰色模型,将强关联变量的降频数据作为多变量离散灰色模型的驱动因素,从而构建基于网络搜索信息的农村水环境质灰色预测模型。实例分析结果表明,相对于传统灰色模型,引入网络搜索信息可以提高农村水环境预测精度,为农村水环境治理提供决策支持。     

SELECTING PARAMETERS FOR SHORT-TERM FORECASTING TECHNIQUES

The objective of this paper is to discover which of three forecasting modes used to select parameters for four short-term forecasting techniques minimizes errors. The study also examines whether the amount of historical data used to find parameters contributes to forecasting success. The results show the traditional one-ahead search routine works well in some, but not all, forecasting situations. Also, forecasting errors appear to decline when more historical data are included in the parameter search.     

基于竞争型制造商创新投入的零售商需求预测信息共享研究

本文考虑两个制造商的竞争,从创新投入的视角构建零售商需求预测信息共享模型,运用贝叶斯统计理论和斯坦伯格博弈方法,探讨了制造商之间的竞争和其开展的成本降低创新对零售商需求预测信息共享的影响。研究发现:(1)制造商的创新投入决策受竞争对手和零售商信息共享的双重影响。(2)由于两个竞争型制造商进行成本降低创新,存在一定的条件使得零售商可以通过免费共享需求信息而获利。(3)当制造商们的创新能力较强时,供应链能够自发达到完全信息共享状态。当制造商们的创新能力较弱时,零售商的信息共享价值为负,竞争型制造商可以通过支付信息共享费用来激励零售商共享需求预测信息。制造商支付的信息共享费用与创新能力、上游竞争强度、预测准确性和随机需求波动性正相关。本文的研究克服了Shang等(2016)关于上游竞争型供应链中零售商信息共享研究中未考虑制造商创新的不足,并进一步探讨了上游竞争型供应链中的信息共享激励机制。     

考虑成分股联跳与宏观信息发布的沪深300指数已实现波动率模型研究

利用日内高频数据计算的已实现波动率较好度量了金融资产的风险,因此对其预测模型的研究具有重要意义。考虑到指数成分股的联跳可能蕴含指数跳跃所未能反映的信息,提出运用非参数方法识别指数成分股的联跳,采用自回归条件风险模型估计成分股联跳强度,并将其引入指数的已实现波动率异质自回归（HAR-RV-CJ）模型中,分析模型预测性能的改进。进一步的,考虑到宏观信息公告的发布可能对股市产生整体性影响,相应影响成分股联跳的几率;因此,在成分股联跳的自回归条件风险模型中引入居民消费价格指数、国内生产总值、贸易差额等宏观信息公告变量,并分析对联跳强度估计以及指数已实现波动率预测的影响。采用2011年1月4日至2013年7月11日沪深300指数及其成分股高频数据的实证表明,指数成分股联跳与指数跳跃具有不同的特征;用成分股联跳强度代替HAR-RV-CJ模型中的跳跃构建的HAR-RV-CI模型,较原始的HAR-RV-CJ模型,以及同时考虑指数跳跃与成分股联跳强度的HAR-RV-CJI模型,具有明显较优的样本内拟合与样本外预测性能。引入宏观信息公告变量可以改进联跳强度自回归条件风险模型的拟合效果,并提高指数已实现波动率模型的样本内拟合能力,但对于指数已实现波动率的样本外预测性能并无明显的帮助。     

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。     

Improving the accuracy and operational predictability of project cost forecasts: an adaptive combination approach

Conducting an early warning forecast to detect potential cost overrun is essential for timely and effective decision-making in project control. This paper presents a forecast combination model that adaptively identifies the best forecast and optimises various combinations of commonly used project cost forecasting models. To do so, a forecast error simulator is formulated to visualise and quantify likely error profiles of forecast models and their combinations. The adaptive cost combination (ACC) model was applied to a pilot project for numerical illustration as well as to real world projects for practical implementation. The results provide three valuable insights into more effective project control and forecasting. First, the best forecasting model may change in individual projects according to the project progress and the management priority (i.e. accuracy, outperformance or large errors). Second, adaptive combination of simple, index-based forecasts tends to improve forecast accuracy, while mitigating the risk of large errors. Third, a post-mortem analysis of seven real projects indicated that the simple average of two most commonly used cost forecasts can be 31.2% more accurate, on average, than the most accurate alternative forecasts.     

Combining forecasts of electricity consumption in China with time-varying weights updated by a high-order Markov chain model

Electricity consumption forecasting has been always playing a vital role in power system management and planning. Inaccurate prediction may cause wastes of scarce energy resource or electricity shortages. However, forecasting electricity consumption has proven to be a challenging task due to various unstable factors. Especially, China is undergoing a period of economic transition, which highlights this difficulty. This paper proposes a time-varying-weight combining method, i.e. High-order Markov chain based Time-varying Weighted Average (HM-TWA) method to predict the monthly electricity consumption in China. HM-TWA first calculates the in-sample time-varying combining weights by quadratic programming for the individual forecasts. Then it predicts the out-of-sample time-varying adaptive weights through extrapolating these in-sample weights using a high-order Markov chain model. Finally, the combined forecasts can be obtained. In addition, to ensure that the sample data have the same properties as the required forecasts, a reasonable multi-step-ahead forecasting scheme is designed for HM-TWA. The out-of-sample forecasting performance evaluation shows that HM-TWA outperforms the component models and traditional combining methods, and its effectiveness is further verified by comparing it with some other existing models.     

基于动态模型平均的大豆期货价格预测研究

针对大豆期货价格波动的复杂性及影响因素的多元性,本文将动态模型平均理论引入大豆期货价格分析与预测研究中,通过动态选择解释变量和系数时变程度,在有效控制模型和系数不确定性的同时,最大限度综合利用大豆期货市场内外部信息,以提高大豆期货价格预测准确度。具体的,本文提出一套基于动态模型平均理论的大豆期货价格影响因素与预测分析框架,从期货市场和经济环境等两方面准确地识别出大豆期货价格影响因素的时变特征,进而构建大豆期货价格预测模型,并通过预测误差指标和Diebold-Mariano检验法评估其与基准模型的预测能力。研究结果表明,动态模型平均理论在有效剖析大豆期货价格影响因素的时变特征的同时,能明显提升大豆期货价格预测准确度。     

二元GED-GARCH模型的利率与汇率波动溢出效应研究

分别运用二元N-GARCH模型和二元GED-GARCH模型,对金融危机前后利率和汇率的波动溢出效应进行研究,通过自适应绝对偏差和自适应均方误差的平方根2种标准进行评价。研究认为,二元GED-GARCH预测效果更好,在金融危机前利率与汇率之间存在着由汇率到利率的溢出效应;在金融危机之后,利率与汇率具有双向的波动溢出效应。     

A COMPARATIVE ANALYSIS OF THE PREDICTIVE ABILITY OF ADAPTIVE FORECASTING,RE-ESTIMATION,AND RE-IDENTIFICATION USING BOX-JENKINS TIME-SERIES ANALYSIS ON QUARTERLY EARNINGS DATA

This paper assesses the predictive ability of the Box-Jenkins methodology when utilized in an on-going setting. Three procedures are utilized to update the original forecasts generated from the Box-Jenkins models: adaptive forecasting, re-estimation, and re-identification. The results indicate that constant monitoring of the structure and parameters of the time-series models are necessary through time. It appears that adaptive forecasting techniques are insufficient to update BJ time-series models when used in conjunction with quarterly earnings data. Re-estimation is recommended as each new observation becomes available. Re-identification procedures are recommended on a less frequent basis.     

面向灾害应急物资需求的灰色异构数据预测建模方法

多源信息集结对提高自然灾害环境下统计数据可信度具有重要作用,但信息渠道的多源性极易导致集结信息数据类型不一致、不兼容,形成灰色异构数据序列。本文应用灰色系统建模技术对灰色异构数据预测建模方法展开研究,首先,基于"核"和"灰度"对灰色异构数据进行规范化处理;然后,建立灰色异构数据"核"序列的DGM(1,1)模型,并以"核"为基础,根据灰度不减公理,以灰色异构数据序列中最大灰度值所对应的信息域作为预测结果之信息域,推导并构建了灰色异构数据预测模型;最后,将该模型应用于某地震帐篷需求量的预测。本文研究成果将传统灰色模拟及预测模型建模对象从"同质数据"拓展至"异构数据",对丰富与完善灰色模拟及预测模型理论体系,提高自然灾害救援效率具有积极意义。     

多因素数据重心预测方法及应用研究

本文基于数据重心概念,通过大量的研究和推导,提出数据重心参数估计理论,并利用数据重心法估计多项式回归预测模型的参数,应用于我国钢材消费量的预测。从应用的结果看,增加了我国目前钢材预测的方法。本方法能最大限度地平滑预测模型的误差,而且应用条件比最小二乘法宽松;也不会因为个别残差较大的异常点而对预测结果产生不稳定性,从而提高了拟合和预测的稳健度,计算更简捷。     

基于向量夹角余弦的组合预测模型的性质研究

基于向量夹角余弦的组合预测是一种相关性的组合预测模型,它是研究组合预测方法的一个新途经.针对基于向量夹角余弦准则下组合预测模型,研究它的基本结构特征.首先提出新的优性组合预测、预测方法优超、冗余度等概念.然后探讨了非劣性组合预测、优性组合预测以及冗余预测方法的存在性,并给出冗余信息的判定定理.最后进行实例分析,表明该方法有较大的实际应用价值.     

AN EMPIRICAL EVALUATION OF CONSTANT AND ADAPTIVE COMPUTER FORECASTING MODELS FOR INVENTORY CONTROL

This study compares the performance of one constant model with two adaptive models. MAD and bias were used as criteria for evaluating performance. The results of testing the adaptive models with the constant model were contrary to the expected better performance of the adaptive models; no significant differences were found between the models over known horizontal, trend, and seasonal demand patterns. However, some of the practical advantages of adaptive forecasting models in business suggest their continued use.     

基于漂移度的组合预测方法研究

由于不同的预测方法能够提供不同的有用信息,其预测精度往往也存在差异,为了分散预测的风险,采用组合预测方法。本文首先提出相容方法集和互补模型集,然后在对不同单一预测模型的漂移性和互补性研究的基础上提出了基于漂移度的组合预测模型,为组合预测模型研究提供一种新的思路。最后通过实例来说明基于漂移度的组合预测模型能够提高样本期预测精度和外推预测精度及实际应用的有效性。