A DISTANCE MEASURE FOR CLASSIFYING ARIMA MODELS

Abstract. In a number of practical problems where clustering or choosing from a set of dynamic structures is needed, the introduction of a distance between the data is an early step in the application of multivariate statistical methods. In this paper a parametric approach is proposed in order to introduce a well-defined metric on the class of autoregressive integrated moving-average (ARIMA) invertible models as the Euclidean distance between their autoregressive expansions. Two case studies for clustering economic time series and for assessing the consistency of seasonal adjustment procedures are discussed. Finally, some related proposals are surveyed and some suggestions for further research are made.     

Composition and seasonal variation of fatty acids of <Emphasis Type="Italic">Diplodus vulgaris</Emphasis> L. from the Adriatic Sea

Muscle tissue from the common two-banded sea bream Diplodus vulgaris L. originating from the Adriatic Sea, Croatia, was analyzed. The FA composition of neutral (TAG) and polar (PE, PC, PI/PS) lipid classes was determined, as well as the lipid and water contents during winter and summer periods. Both the total lipid and water contents were higher in the winter period. We identified 16 different FA. The major constituents of the total FA in both seasons were saturates: palmitic (16∶0) and stearic acids (18∶0); monoenes: oleic (18∶1n−9) and palmitoleic acids (16∶1n−7); and polyunsaturates: arachidonic acid (20∶4n−6), EPA (20∶5n−3), and DHA (22∶6n−3), but their amounts and ratios differed significantly between the two seasons and between lipid fractions. The FA composition showed a noticeable pattern of seasonality that reflected fluctuations mainly in TAG. The diminution of the monounsaturated FA content in the summer was clearly followed by an increase in PUFA content. Diplodus vulgaris is a good source of natural n−3 PUFA and would therefore be suitable for inclusion in highly unsaturated low-fat diets.     

太阳能跨季节储热耦合热泵系统性能影响因素分析

通过建立太阳能跨季节储热耦合热泵系统物理模型和非稳态传热数学模型,利用VisualBasic软件对太阳能跨季节储热耦合热泵系统性能进行了仿真模拟,并对集热器循环水泵启动温度、储热时间段、水箱容积与集热器面积比(V/A)和地下储热循环水泵启动温度等因素对系统性能的影响进行了模拟计算与分析.研究结果表明,在进行太阳能跨季节储热耦合热泵系统设计和系统运行性能分析时,需要综合考虑上述因素的优化配置,以此才能保证系统运行可靠、经济和节能.     

快递企业人员调配问题的建模与求解方法

针对快递企业配送中心人员调配问题,提出了在服务于同类客户与不同类客户两种条件下,用时间序列季节系数法预测来件量,并分析快件处理和配送过程中的人员、车辆因素的方法.该方法对同类客户调整配送中心快件处理人员与分配人员的数量,以此达到极小化快件剩余量的目标,对不同类客户调整配送人员的数量,以此达到极小化配送中心成本的目标,并建立了数学模型.利用粒子群算法对模型进行求解,通过实例进行仿真试验,从而得到配送中心人员调配的最佳方式,以此指导企业合理运营.     

贝叶斯方法的污染源季节性排放量控制和管理

为保证水质功能区内水质达标,必须对区域内污染源排放量进行控制和管理。选取松花江哈尔滨段主要污染指标COD和氨氮为研究对象,利用一维水质模型构建各污染源排放量与河流水质之间的关系。用贝叶斯方法估计水质模型中的重要参数综合降解系数（ k）,根据估计时期的不同,分别建立季节模型和年度模型,用以控制既定水质目标下各污染源排放量。结果表明,季节模型的预测效果较好,能更好地表达水中污染物的综合降解浓度。应用季节模型和贝叶斯方法开展季节性水质管理工作,可以提供给决策者更多的信息,有助于对污染源排放量的不确定性进行量化和评估。此外,通过对比各污染源的削减水平,可得各污染源在不同时期的控制权重,从而使管理者在不同时期有针对性地对污染源排放量进行控制。贝叶斯方法在污染源排放量控制中的应用可以增强水质模型的预测能力,有效提高水质管理的水平。     

渭河干流关中段季节性水环境容量研究

针对季节性河流流量具有周期性变化的特点,采用季节性污水排放计划(SDP)的季节划分方法,提出季节性河流水环境容量计算方法。以渭河干流关中段为例,将一年划分为旱雨两季,利用一维水质模型计算8个季节划分方案和1个对照方案在50%、75%和90%保证率下的水环境容量。研究结果表明:在相同保证率下,分季节计算的渭河干流关中段COD和NH3-N环境容量分别是传统环境容量计算方法的1.17~1.59倍和1.25~1.79倍,表明分季节计算的河流水环境容量明显大于以年为单位计算的河流水环境容量。     

近100多年来香港和澳门气温时序变化特征及其与ENSO的关联性

基于近 100 多年来的均一化气温数据,从趋势变化、季节差异、突变特征、周期振荡方面诊断香港和澳门的气温变化特征。结果表明: ( 1) 港澳年平均气温和不同季节气温在近 100 多年来均呈显著增加趋势,其中春冬季节增加速率较大,夏秋季节增加速率较小。除秋冬季节外,其它时间上的香港气温增加趋势均大于澳门。港澳夏冬季节气温差异均呈减少趋势。( 2) 香港和澳门年平均气温分别于 1963 年和 1991 年发生突变。在季节上,港澳地区的气温突变均在秋季发生年份最晚,冬季次之,夏季再次之,春季最早。( 3) 香港年平均气温分别存在 3 a、7 a、15 a 和 30 a、60 a 和 120 a 的周期振荡特征。澳门年平均气温存在 3 a、6 a、12 a、24 a 和 48 a 左右的周期振荡特征。( 4) ENSO 与香港年平均气温在 4 a 以下尺度以同位相或提前 1 a 位相变化关系为主,在 4 a 以上尺度以同位相变化关系为主。ENSO 与澳门年平均气温在不同时频域均以同位相变化关系为主。     

淮安市地表水水质变化趋势分析

应用季节性kendall检验法,对2000—2009年淮安市主要水系河流的地表水水质指标变化趋势进行分析评价。结果表明:随着近年来淮河流域水污染防治力度的加大,以及对洪泽湖作为南水北调东线工程输水主干线自身的治理,淮安市境内的主要河流湖泊总体水质略有好转,由于淮河入海水道水功能区划的调整,入海水道总体水质有所下降,而洪泽湖水体TP含量呈逐年上升趋势,其他水体水质总体变化不大。     

时间序列季节调整方法在中国的发展:PBC版X-12-ARIMA

中国人民银行(PBc)版X-12-ARIMA软件是基于中国特点而定制的时间序列季节调整软件.通过总结时间序列季节调整方法的特点以及相应软件在国外的发展,针对我国应用的特点,尤其是春节因素的考虑,在解剖X-12-ARIMA方法原理的基础上,在春节因素计算方法、软件应用界面以及用户使用帮助等3个主要方面加以改进,具有数据导入、调整设置文件、运行方式以及结果输出4方面的特色.     

Foliage constituents of douglas fir (Pseudotsuga menziesii (Mirb.) Franco (Pinaceae)): Their seasonal variation and potential role in douglas fir resistance and silviculture management

Seasonal changes in the production of primary nutrients (soluble carbohydrates) and secondary metabolites (terpenes, monomeric phenolics, and tannins) in the current needle tissue of Douglas fir were investigated. All four classes of compounds showed significant seasonal changes in concentration during foliage development. Most terpenes increased significantly in concentration from June 11 to August 3, and then showed declining concentrations to September 20. The most dramatic and significant seasonal increases occurred in-pinene, camphene, and bornyl acetate concentrations. The monomeric phenolics chlorogenic acid, taxifolin glucoside, quercetin galactoside, and those unknown phenolics showed an overall trend of declining in concentration from June 11 through September 20. However, considerable variation between sampling dates in the concentration of these phenolics was noted. Tannin concentration decreased significantly from June 11 to July 9, and then increased in concentration to the September 20 sampling date. Fructose, galactose, glucose, and sucrose tended to decrease from June 11 to September 20. However, significant variation between sampling dates was evident in these compounds as well. Galactose was the major compound in the soluble carbohydrate fraction, amounting to almost 80% of the total concentration throughout the growing season. These data suggest that if phenolics and tannins function as defenses, they would only affect second- and possibly third-instar budworm larvae during the time that these instars mine the buds. Camphene,-pinene, and bornyl acetate increased in concentration throughout the growing season and may be effective deterrents to the budworm. Both bornyl acetate and camphene have been shown in field and laboratory studies to increase larval mortality and adversely affect budworm larval growth. Carbohydrates generally act as nutrients that enhance herbivore growth. However, in a previous study, galactose was found to cause reduced budworm larval growth and increased larval mortality.