Comparison of forecasting methods with an application to predicting excess equity premium

This paper reviews various forecast methods including combination using theoretically optimal weights and those under model selection approaches. In addition, we suggest two modified simple averaging forecast combination methods—a mean corrected and a mean and scale corrected method. We conclude that due to the fact that real data is usually subject to structural breaks, rolling forecasting scheme has a better performance than fixed window and continuously updating scheme. In addition, methods that use less information appear to perform better than methods using all the sample information about the covariance structure of the available forecasts. The mean and scale corrected simple average approach yield smaller mean squared forecast error than the three widely used regression approaches suggested by Granger and Ramanathan [11].     

Effect of aspect ratio on convection in a porous enclosure with partially active thermal walls

The aim of the present numerical investigation is to understand the effect of aspect ratio and partially thermally active zones on convective flow and heat transfer in a rectangular porous enclosure. Five different heating and cooling zones are considered along the vertical walls while the remaining portions of the sidewalls and top and bottom of the enclosure are adiabatic. The Brinkman-Forchheimer extended Darcy model is used in the study. The governing equations are solved by the finite volume method with the SIMPLE algorithm. The computations are carried out for a wide range of parameters and the results are presented graphically. The results reveal that the location of heating and cooling zones has a significant influence on the flow pattern and the corresponding heat transfer in the enclosure. The rate of heat transfer approaches to a constant value for very low values of the Darcy number. The heat transfer rate is decreased on increasing the aspect ratio.     

一种使用分量组合的彩色图像压缩方法

为了适应图像的存储和传输,需要对图像进行压缩以减少数据量。对于彩色图像而言,其色彩模式和数据组合直接影响了压缩效率,压缩时通常采用每个分量单独压缩后再统一编码,未考虑分量之间的组合关系。根据彩色图像的数据结构将图像的RGB分量排列组合成新的灰度图像,将其中一个分量作为基准实行粗细分层操作,汇同该分量与其他两个分量的差值图像分别按JPEG压缩标准进行压缩编码。实验表明,该算法得到的压缩图像在与JPEG同等质量的条件下,数据量要小于JPEG压缩的数据量,耗时短,同时可为快速的彩色图像搜索和预览提供技术参考。     

一种用于微喷的感应加热器设计与仿真

提出了一种用于热微喷头的感应加热器新结构。进行了相关的结构和工艺设计,在MEMS工艺环境下制作出了可以非接触加热,工艺结构简单,使用寿命长的微感应加热器。使用ANSYS对微感应加热器进行了建模和仿真分析,结果显示:感应加热器的最大加热功率可达到5 mW以上,满足设计要求,同时得到了感应加热器的功率影响因素。     

Three-objective fuzzy chance-constrained programming model for multiproject and multi-item investment combination

This paper presents a new fuzzy chance-constrained programming model to find the solution for multiproject and multi-item investment combination in investment combination problems. The proposed 0-1 integer programming model has three objectives with fuzzy constraints, and NSGA-II is applied to solve the optimization model with a small modification of the constraint-handling rule. A simulation experiment illustrating the application of the proposed model is presented and Pareto-optimal solutions are obtained through a modified NSGA-II algorithm. A comparison among NSGA-II, PSO, and DE shows that modified NSGA-II has some advantages over PSO and DE.     

Combining feature spaces for classification

In this paper we offer a variational Bayes approximation to the multinomial probit model for basis expansion and kernel combination. Our model is well-founded within a hierarchical Bayesian framework and is able to instructively combine available sources of information for multinomial classification. The proposed framework enables informative integration of possibly heterogeneous sources in a multitude of ways, from the simple summation of feature expansions to weighted product of kernels, and it is shown to match and in certain cases outperform the well-known ensemble learning approaches of combining individual classifiers. At the same time the approximation reduces considerably the CPU time and resources required with respect to both the ensemble learning methods and the full Markov chain Monte Carlo, Metropolis-Hastings within Gibbs solution of our model. We present our proposed framework together with extensive experimental studies on synthetic and benchmark datasets and also for the first time report a comparison between summation and product of individual kernels as possible different methods for constructing the composite kernel matrix.     

基于多种视频特征的镜头边界检测算法

针对固定场景的视频,提出一种基于时空分割的视频分割算法.该算法在时域中利用Tophat形态学滤波得到视频对象的精确位置,在空域中采用基于t混合模型和贪婪EM的聚类算法进行单帧图像分割.将时域定位和空域分割结果结合,可以准确地将视频序列中感兴趣的运动目标分割出来.实验表明,该算法能够得到完整的视频对象,有一定的理论意义和实用性.     

四自由度并联跟踪台的软件关键技术

研究四自由度并联跟踪台中的软件关键技术,编写Matlab仿真程序对并联跟踪台进行参数化设计,利用Solidworks与ADAMS联合仿真软件获取并联跟踪台的运动学和动力学性能参数,通过对操纵杆的编程实现对并联跟踪台的手动控制。实验结果表明,四自由度并联跟踪台设计合理、运动灵活、手控效果好。     

交互式凸组合法的混合时滞不确定中立系统的稳定性

针对带分布时滞和离散时滞的不确定中立型系统进行稳定性研究. 基于交互式凸组合方法和下界引理, 通过构造恰当的李雅普诺夫泛函, 适当分割时滞区间, 处理一组由凸参数逆加权的正函数线性组合(交互式凸组合), 给出线性矩阵不等式形式的系统鲁棒稳定性判据. 该方法允许离散时滞为变时滞, 增强了系统的鲁棒性能. 数值算例验证了所得结果的有效性和合理性.

     

谈话节目语料库的构建与会话结构分析

口语语料库的建设是口语研究的基础工作,该文选择具有代表性的交谈式谈话节目《锵锵三人行》和对谈式谈话节目《鲁豫有约》作为语料,建立了一个小型的谈话节目语料库,并构建了包含五大类16小类的会话结构标注体系,对语料进行了会话结构的标注。统计得到打断结构309例,插入结构141例,重复结构111例,问答结构653/589例,阻碍—修正结构51/21例,反映了会话结构在数量上的不均衡分布,节目的形式、性质以及交际任务是会话结构分布的主要影响因素。会话结构组合具有模式性,该文使用Trigram方法对其组合情况进行了分析,发现语料中的高频组合是问答毗邻对,此外有大量的非毗邻性组合。会话结构组合模式不但反映出谈话节目的风格特点,还有助于分析会话中的功能性模块、会话策略的形成,进而更加深入地了解会话的运作机制。