基于ARIMA的居民消费价格指数建模与预测

根据我国2007～2010年的实际居民消费价格指数，建立了基于ARIMA的物价指数预测模型。实验结果表明，该模型的绝对误差以及百分比绝对误差都控制在了一定范围之内，因此该模型拟合效果较好，预测值接近实际值。最后，应用该模型对我国2011年1月至5月的居民消费价格指数进行了预测。     

电熔镁砂企业生产计划多目标优化

针对大部分电熔镁砂生产企业都是产能有限的小工厂，在销售旺季时会损失超过自身产能的订单；在销售淡季时又容易积压库存的问题，提出了基于需求预测的多目标生产计划优化方法.该方法利用基于季节性自回归求和移动平均（ARIMA）模型向前多步预测市场的需求量，并建立销量与库存累积量的多目标优化模型.设计了关键月度产量计划后推法求解模型得到了决策者满意的优化目标.对预测过程中存在的偏差，利用实时反馈得到的真实销售数据修正预测结果，进而调整优化模型，校正后期产量计划.实际应用效果证明，按所提方法得到的产量计划能够实时地反映市场需求的动态变化，并实现了库存与销量的多目标优化.
     

面向多媒体取证的电网频率建模

电网频率建模旨在对其随机波动特性建立数学模型，形成特征描述、统计分析以及轨迹预测能力，是基于电网频率判据的数字取证中一项重要的研究子课题。该文首先揭示了现有自回归模型的不足，并基于该模型对实际观测得到的电网频率参考信号数据集进行统计分析，分别用差分整合移动平均自回归（autoregressive integrated moving average,ARIMA）模型和两种基于马尔科夫链的模型进行建模，进而求解模型各项参数。对比实验表明： ARMA (2,4)模型在理论上对华中电网频率具备最优的描述性能，而马尔科夫链模型因考虑到实际场景下频率分辨率的局限性，更适用于模拟从待测文件中提取的电网频率。     

卫星导航系统与气压高度计的信息处理与融合

提出一种卫星导航系统与气压高度计的高度数据融合算法. 利用ARIMA(autoregressive integrated moving average)模型信号降噪技术对卫星导航系统和气压高度计的高度信号进行降噪处理. 建立高度测量系统的状态方程和量测方程,并结合卡尔曼滤波对高度数据进行第1 次融合,再利用递推加权最小二乘法对高度数据进行第2 次融合,得到精度更高的飞行高度信息. 实测数据分析结果表明：该算法所获得的高度信息与气压高度计、卫星导航系统相比,标准差降低一半以上,有效地提高了测量精度,可满足飞行器对高度测量信息的要求.     

联合经验正交分解和ARIMA模型的中国地区电离层短期预报

针对电离层垂直总电子含量(Vertical Total Electron Content,VTEC)具有非平稳和季节性变化的特性,结合经验正交分解(EOF)能够对非平稳时间序列进行简化和剔除冗余信息的优势,该文探索联合EOF和自回归移动平均模型(ARIMA)作为FOE-ARIMA,对中国地区电离层VTEC进行短期预报。采用IGS(International GNSS Service)中心提供的中国地区电离层格网数据(Global Ionospheric Maps,GIM),对不同季节前10天GIM数据进行EOF分解,使用ARIMA模型对主分量进行预报,通过重构获取未来5天VTEC值,并将EOF-ARIMA模型预报结果与ARIMA模型进行对比、分析。结果表明：EOF-ARIMA模型平均相对精度为83.3%,平均标准差为3.51 TECu,较ARIMA模型其平均相对精度提高了3.3%,平均标准差降低了0.16 TECu;EOF-ARIMA模型预测结果无明显季节差异,ARIMA模型秋季预报精度明显低于其它季节;EOF-ARIMA模型在赤道异常处预报精度未受影响。由此表明EOF-ARIMA模型在中国地区进行电离层短期预报具有较高的精度和稳定性。     

The role of time series analysis in the evaluation of econometric models

This paper compares the properties of a structural model—the London Business School model of the U.K. economy—with a time series model. Information provided by this type of comparison is a useful diagnostic tool for detecting types of model misspecification. This is a more meaningful way of proceeding rather than attempting to establish the superiority of one type of model over another. In lieu of a better structural model, the effects of inappropriate dynamic specification can be reduced by combining the forecasts of both the structural and time series models. For many variables considered here these provide more accurate forecasts than each of the model types alone.     

The accuracy of extrapolation methods; an automatic box–jenkins package sift

Hill and Woodworth (1980) proposed an algorithm suitable for identifying Box–Jenkins models automatically without reliance on the investigator. This paper first reviews the method. It is then used on the 111 series analysed by Anderson in the Makridakis forecasting competition. The results show that the automatic method of Hill and Woodworth is comparable in terms of accuracy to the full Box–Jenkins identification procedure.     

The role of forecasting in corporate strategy: The xerox experience

This paper presents the system of analysis used by the Xerox Corporation to relate the external environment to company decisions. The system is sophisticated and elaborate, comes to grips with such issues as product forecasting, market monitoring, activity monitoring, materials and labour cost analysis, and product price analysis. In addition, the system examines the longer-term issues associated with corporate strategy, with the more recent initiatives directed toward the strategic focus The Xerox case illustrates very well how externally provided forecasts of economic environments, both at home and abroad, can be used as inputs to a variety of econometric products to serve the individual corporation. The challenge in this work is to build the bridges from the external forces to the critical company decisions. That is a task which requires sophisticated tools and skilled professionals to accomplish. This case study shows what can be done.     

我国交通货物运输量的时间序列分析

研究我国交通货物运输量与国内生产总值之间的协整关系,建立了货运量、货物周转量、国内生产总值(GDP)的ARIMA模型,对三者之间的Granger因果关系进行了研究,并建立了向量自回归模型.结果显示,交通货物运输量与国民生产总值之间没有协整关系,货物周转量是货运量、国内生产总值的Granger原因,但货运量、国内生产总值不是货物周转量的Granger原因.     

Differential AR algorithm for packet delay prediction

Different delay prediction algorithms have been applied in multimedia communication, among which linear prediction is attractive because of its low complexity. AR (auto regressive) algorithm is a traditional one with low computation cost, while NLMS (normalize least mean square) algorithm is more precise. In this paper, referring to ARIMA (auto regression integrated with moving averages) model, a differential AR algorithm (DIAR) is proposed based on the analyses of both AR and NLMS algorithms. The prediction precision of the new algorithm is about 5?10 db higher than that of the AR algorithm without increasing the computation complexity. Compared with NLMS algorithm, its precision slightly improves by 0.1 db on average, but the algorithm complexity reduces more than 90%. Our simulation and tests also demonstrate that this method improves the performance of the average end-to-end delay and packet loss ratio significantly.