基于小波随机耦合模型的三江平原年降水量预测

近些年来,由于水田面积迅速增加,导致三江平原井灌水稻区地下水位普遍下降,"吊泵"、局部超采现象时有发生.为解决上述问题,以853农场为例,对实测年降水资料进行差分和标准化处理,采用小波理论和随机水文学方法进行耦合分析,建立了年降小波随机耦合模型,精度检验结果表明,模型有效性和可靠性较高.该模型揭示了区域年降水量的时间变化规律,为853农场乃至整个三江平原充分利用天然降水、合理制定水稻灌溉制度以及地下水资源的恢复和可持续利用提供了科学依据.     

组合时间序列ARMA模型在经济预测中的应用——内蒙古十一五期间GDP预测

2006—2010年是内蒙古"十一五"规划的重要时期.利用组合ARMA时间序列模型,对内蒙古2006—2010年GDP进行预测,得出的结果能够帮助我们把握"十一五"期间经济运行的变动趋势,并寻求最佳的调控办法.     

Characterizing the frequency of orders received by a stockist

John E. Boylan Applied Social Sciences and Humanities, Buckinghamshire Chilterns University College, Buckinghamshire HP11 2JZ, UK F. R. Johnston Kerridge Computer Company, Charnham Park, Hungerford RG17 0YU, UK Email: estelle.shale{at}warwick.ac.uk Received on 26 January 2007. Accepted on 19 November 2007. It is regularly asserted that the frequency of orders receivedby a stockist can be represented as a Poisson process, but verylittle corroborative evidence has been published. This paperpresents some results supporting the assumption. An adaptationto the standard testing methodology is presented which overcomesthe complications arising from the necessity of sampling fromtime series data. The new approach is applied to a large rangeof stock-keeping units. Except for the extreme tail of the distribution,which is of little interest for most inventory applications,it is found that the frequency of orders is well representedby a Poisson process.     

Demand forecasting adjustments for service-level achievement

John E. Boylan School of Business and Management, Buckinghamshire New University, Bucks, UK Email: a.syntetos{at}salford.ac.uk Received on 25 January 2007. Accepted on 3 October 2007. Demand forecasting and stock control are traditionally examinedas independent of each other. Even though this weakness hasbeen highlighted in the academic literature, little empiricalwork has been conducted on forecasting adjustments addressingthe interaction between forecasting and stock control. In thispaper, the relevant literature is critically reviewed. Subsequently,the empirical performance of some modifications and adjustments,on slow-moving items, is examined in detail. The data set consistsof the individual demand histories of 753 intermittent lineitems from the Royal Air Force (UK). Overall, the results indicatethat there is a scope for improving the performance of parametricstock control systems, and adjustments are indeed required inorder to account for the interaction between forecasting andstock control.     

VaR methods for the dynamic impawn rate of steel in inventory financing under autocorrelative return

This paper proposes the way of setting the dynamic impawn rate by dividing the impawn periods into different risk windows. In an efficient financial market, the return is hypothetically independent, while in a pledged inventory market where spot transactions predominate, the return is auto-correlative. Therefore, the key to setting the impawn rate is to predict the long-term risk. In this experiment, using the database of spot steel, we established a model with the formula AR (1)-GARCH (1,1)-GED, forecasting the VaR of steel during the different risk windows in the impawn period through a method of out-of-sample, and got the impawn rate according with the risk exposure of banks. The results of our experiment indicated that the introduction of coefficient K into the model can significantly improve bank risk coverage and reduce its efficiency loss. Besides, the impawn rate obtained by the model correlates positively with the lowest price in the future risk windows.     

基于灰色系统的支持向量回归预测方法

根据部分时间序列数据贫信息、高噪声和非线性等特点,采用含边值修正的灰色模型进行预测,获取残差序列后运用支持向量回归（SVR）方法对模型进行残差修正得到复合的灰色支持向量回归模型．在支持向量回归中构造具有自适用性的动态惩罚参数G替代传统SVR中的不变参数来提高模型的准确性,同时构造算法决定￡以平滑过度调节．广东省工业生产指数的预测试验结果表明,复合模型具有比其他简单模型更理想的预测效果．     

随机时间序列法在三门峡潼关站年来水量预测中的应用

以时间序列分析法为理论依据,建立了三门峡潼关站年来水量的预测模型,预测检验结果表明,该模型用于三门峡潼关站年来水量的趋势预测,效果良好.     

陕西省城镇居民收入结构的灰色关联分析与趋势预测

应用灰色关联分析方法,分两个历史阶段对比研究了不同收入来源对陕西城镇居民人均可支配收入的影响,结果表明,居民收入结构不合理和发展不平衡现象较为严重,工薪收入和转移性收入是影响陕西居民收入的主要因素,并且其重要程度有进一步增强的趋势,其他来源的收入对居民收入增长有明显的抑制作用.最后,通过建立GM(1,1)新陈代谢模型群,对陕西城镇居民未来几年的人均可支配收入和工薪收入进行了预测.     

基于灰色神经网络的企业风险特征指标动态预测方法研究

根据企业风险特征指标预测问题的特点,提出将灰色系统GM(1,1)模型与神经网络结合建立一阶灰色神经网络预测模型,以实现系统预测的动态性及提高系统的预测精度.但该模型具有一定的局限性,从模型参数的角度给出了该模型只适用于具有"单调"性数据的证明,进而提出了三阶灰色神经网络预测模型,以适应预测数据"非单调"或摆动的情况.但随着系统建模过程中阶数的增加,预测精度会有所下降,因此应根据数据特点选择预测模型.最后,通过实证分析验证了上述模型及证明结论.     

基于灰色聚类-SPA的建筑施工项目安全风险评价

针对建筑施工阶段的安全风险,建立了基于灰色聚类-集对分析(SPA)的建筑施工项目安全风险的综合评价模型.在应用SPA确定安全风险评价因素权重的基础上,采用灰色聚类方法对建筑施工项目进行聚类分析,评价建筑施工项目安全风险所处的等级,最后依据SPA中集对势等级次序表分析建筑施工项目安全风险发展态势.实例分析证明了该模型的适用性和合理性,对比分析表明其评价结果的可靠性,为施工企业的安全风险决策提供借鉴.