基于遗传算法和熵的缩减记忆式LS-SVM财务困境预测模型研究

为了提高财务困境预测的正确率,减少模型的训练样本数和训练时间,在传统支持向量机(SVM)预测模型的基础上,将遗传算法、信息熵和缩减记忆算法应用于最小二乘支持向量机(LS-SVM),提出了一种基于遗传算法和信息熵的缩减记忆式最小二乘支持向量机预测模型。并独立推导出了适合财务困境预测这一离散序列的熵以及支持向量机核函数的表达式,同时,给出了这一改进模型的实现步骤。实验结果表明,该模型无论是预测正确率,还是训练样本的数量和训练时间,都显著优于最小二乘支持向量机以及传统支持向量机模型。     

果蝇优化最小二乘支持向量机混合预测模型—以我国物流需求量预测为例

为解决最小二乘支持向量机参数设置的盲目性,利用果蝇优化算法对其参数进行优化选择,进而构建了果蝇优化最小二乘支持向量机混合预测模型.以我国物流需求量预测为例,验证了该模型的可行性和有效性.实例验证结果表明:与单一最小二乘支持向量机和模拟退火算法优化最小二乘支持向量机预测模型相比,该模型不仅能够有效选择参数值,而且预测精度更高.     

基于EEMD的LS-SVM和BP神经网络混合短期负荷预测

提出了基于总体平均经验模态分解(EEMD)、最小二乘支持向量机(LSSVM)和BP神经网络的实用综合短期负荷预测方法,进行电力系统短期负荷预测.首先运用EEMD方法将非平稳的负荷序列分解,然后根据分解后各分量的特点选用最佳的核函数,利用最小二乘支持向量机分别对各分量进行预测,最后对各分量预测结果采用BP神经网络重构得到最终的预测结果.对实测数据的分析表明基于该综合方法的电力系统短期负荷预测具有较高的精度.     

天然气消费量的偏最小二乘支持向量机预测

结合偏最小二乘法和支持向量机的优缺点,提出基于偏最小二乘支持向量机的天然气消费量预测模型。首先,利用偏最小二乘法确定影响天然气消费量的新综合变量,建立以新综合变量为输入,天然气消费量为输出的支持向量机模型,对天然气消费量进行了预测;然后,与多元回归、偏最小二乘回归、普通支持向量机做误差检验比较,验证该方法的可行性与正确性。结果表明,此天然气消费量预测模型具有较高的精确度和应用价值。     

电网项目融资租赁信用评价混合模型的新研究

电网建设工程通过项目融资租赁进行快速融资的同时,给租赁公司带来巨大的信用风险.通过事前对承租人进行信用评价,能够有效降低信用风险损失.针对电网企业信用评价的多属性非线性特征,提出了基于独立分量分析技术-支持向量机的信用评价混合模型.首先,采用独立分量分析技术对信用属性数据进行属性重构,实现属性数据的去噪.然后,将重构后的新信用属性数据用于支持向量机的训练建模.最后,通过实例模拟对比分析了独立分量分析技术对支持向量机分类的有效性.结果表明,独立分量分析技术能够改善信用属性数据特征,并且在多属性分类问题中,独立分量分析技术有助于提高支持向量机分类的准确率.     

基于混沌最小二乘支持向量机的时间序列预测研究

为了对这种具有非线性特性的时间序列进行预测,提出一种基于混沌最小二乘支持向量机.算法将时间序列在相空间重构得到嵌入维数和时间延滞作为数据样本的选择依据,结合最小二乘法原理和支持向量机构建了基于混沌最小二乘支持向量机的预测模型.利用此预测模型对栾城站土壤含水量时间序列进行了预测.结果表明,经过相空间重构优化了数据样本的选取,通过模型的评价指标,混沌最小二乘支持向量机的预测模型能精确地预测具有非线性特性的时间序列,具有很好的理论和应用价值.     

基于灰色关联分析与PLS-LSSVM的航材备件需求预测模型

航材备件是保障航空装备日常训练和作战正常使用的重要影响因素,针对部分航材备件样本数据量少,影响因素多且复杂多变,预测结果与装备系统完好性要求偏差较大等问题.建立基于灰色关联分析(GRA)与偏最小二乘(PLS)及最小二乘向量机(LSSVM)相结合的航材备件预测模型,采集某无人机航材备件数据,通过对统计数据进行灰色关联分析,提取航材备件需求的相关因素作为模型训练样本,确定关键因素,利用偏最小二乘对关键因素特征提取,然后将偏最小二乘特征提取后的数据作为最小二乘向量机输入,进行模型构建及分析.通过实验验证了该方法的可行性与适用性,能够满足无人机航材备件预测的实际需要.     

基于核主元分析的带可变惩罚因子最小二乘模糊支持向量机模型及其在信用分类中的应用

信用分类是信用风险管理中一个重要环节,其主要目的是根据信用申请客户提供的资料从申请客户中区分出可信客户和违约客户,以便为信用决策者提供决策依据.为了正确区分不同的信用客户,特别是违约客户,结合核主元分析和支持向量机算法构造基于核主元分析的带可变惩罚因子最小二乘模糊支持向量机模型对信用数据进行了分类处理.在基于核主元分析的带可变惩罚因子最小二乘模糊支持向量机模型中,首先对样本数据进行预处理,然后利用核主元分析以非线性方式降低数据的维数,最后利用带可变惩罚因子最小二乘模糊支持向量机模型对降维后数据进行分类分析.为了验证,选择两个公开的信用数据集来进行实证分析.实证结果表明：基于核主元分析的带可变惩罚因子最小二乘模糊支持向量机模型取得了较好的分类结果,可为信用决策者提供重要的决策参考依据.     

基于LS-SVM的管道腐蚀速率灰色组合预测模型

为提高管道腐蚀速率预测精度,建立了一种基于最小二乘支持向量机的灰色组合预测模型.以各种灰色模型对管道腐蚀速率的预测结果作为支持向量机的输入,以管道腐蚀速率的实测值作为支持向量机的输出,采用最小二乘支持向量机回归算法和高斯核函数对支持向量机进行训练,利用训练好的支持向量机进行组合预测.预测模型兼具灰色模型所需原始数据少、建模简单、运算方便的优势和最小二乘支持向量机具有泛化能力强、非线性拟合性好、小样本等特性,弥补了单一预测模型的不足,避免了神经网络组合预测易于陷入局部最优的弱点.模型结构简单、实用,仿真结果验证了其有效性.     

基于邻域粗糙集属性约简的对偶约束式LS-SVM财务困境预测模型研究

为了提高财务困境预测的正确率,改善模型预测的效果,将邻域粗糙集和遗传算法应用于对偶约束式最小二乘支持向量机,提出了一种基于邻域粗糙集属性约简的对偶约束式最小二乘支持向量机预测模型.同时,给出了这一改进模型的实现步骤.实证结果表明,通过邻域粗糙集指标预处理和遗传算法参数优化后,不但提高了模型预测的正确率,还降低了模型运行的时间,证实了该模型应用于财务困境预测是有效的.     

基于智能化信息处理的建筑工程造价短期预测

提出了一种新的基于智能化信息处理的建筑工程造价短期预测模型.该模型首次利用数据的高阶统计信息,提出了改进独立分量分析技术.通过构建适用于建筑工程造价的属性重构空间,挖掘出表征能力更强的造价独立属性,用于神经网络的学习和训练,从而建立了全新的建筑工程造价短期预测智能模型.该模型通过发挥独立分量分析强大的信号分离能力,增强了神经网络的学习效率,提高了预测精度.实例数据验证了文中所建模型的有效性.     

Efficiency measurement using independent component analysis and data envelopment analysis

Efficiency measurement is an important issue for any firm or organization. Efficiency measurement allows organizations to compare their performance with their competitors’ and then develop corresponding plans to improve performance. Various efficiency measurement tools, such as conventional statistical methods and non-parametric methods, have been successfully developed in the literature. Among these tools, the data envelopment analysis (DEA) approach is one of the most widely discussed. However, problems of discrimination between efficient and inefficient decision-making units also exist in the DEA context (Adler and Yazhemsky, 2010). In this paper, a two-stage approach of integrating independent component analysis (ICA) and data envelopment analysis (DEA) is proposed to overcome this issue. We suggest using ICA first to extract the input variables for generating independent components, then selecting the ICs representing the independent sources of input variables, and finally, inputting the selected ICs as new variables in the DEA model. A simulated dataset and a hospital dataset provided by the Office of Statistics in Taiwan’s Department of Health are used to demonstrate the validity of the proposed two-stage approach. The results show that the proposed method can not only separate performance differences between the DMUs but also improve the discriminatory capability of the DEA’s efficiency measurement.     

A life-cycle optimization model using semi-markov process for highway bridge maintenance

Due to a variety of risks related to aging, construction, material degradation, harsh environment, increasing traffic, and insufficient capacity, a large percentage of bridges in the U.S. highway system are deteriorating beyond acceptable standards. Although significant investments are needed to bring bridges back to acceptable condition, most highway agencies lack the appropriate funding and therefore need effective methodologies for allocating limited resources efficiently and cost-effectively. This paper presents a life-cycle optimization model using a semi-Markov process and demonstrates how the proposed method can assist highway agencies to make more quantitative and explicit decisions for bridge maintenance. The 2012 National Bridge Inventory (NBI) dataset for the State of Texas was analyzed in this study to illustrate bridge structural responses and behaviors under uncertainty and risks. The proposed method is accurate when compared to real data and customized to help highway agencies to optimize their decisions on structuring bridge maintenance, and consequently, leading to cost savings and more efficient sustainability of their bridge systems. The major contribution of this research is the low-error model and process algorithm for selecting the most appropriate maintenance strategy. If employed properly, it may allow agencies to more effectively maintain an aging infrastructure system.     

基于ARIMA和LSSVM的非线性集成预测模型

针对复杂时间序列预测困难的问题,在综合考虑线性与非线性复合特征的基础上,提出一种基于ARIMA和最小二乘支持向量机(LSSVM)的非线性集成预测方法.首先采用ARIMA模型进行时间序列线性趋势建模,并为LSSVM建模确定输入阶数;接着根据确定的输入阶数进行时间序列样本重构,采用LSSVM模型进行时间序列非线性特征建模;最后采用基于LSSVM的非线性集成技术形成一个综合的预测结果.将该方法用于中国GDP预测取得的结果,与单独预测方法及流行的其他集成预测方法相比,预测精度有了较大的提高,从而验证了方法的有效性和可行性.     

Hybrid linear and nonlinear complexity pursuit for blind source separation

Blind source separation (BSS) is an increasingly popular data analysis technique with many applications. Several methods for BSS using the statistical properties of original sources have been proposed; for a famous case, non-Gaussianity, this leads to independent component analysis (ICA). In this paper, we propose a hybrid BSS method based on linear and nonlinear complexity pursuit, which combines three statistical properties of source signals: non-Gaussianity, linear predictability and nonlinear predictability. A gradient learning algorithm is presented by minimizing a loss function. Simulations verify the efficient implementation of the proposed method.     

Optimization techniques for planning highway pavement improvements

In this paper, two different decision models for the planning of highway pavement improvements are presented. In the first model, we want to get a prescribed improvement in the state of the highway network with minimal agency cost. In the second model, a given amount of money is distributed between the highway sections in different states in such a way that the achieved improvements should be the best in some sense. The first model helps the administration in the estimation of the necessary cost for the annual highway improvements. The second model in addition gives an objective tool to the administration for the distribution of the total amount of money between the different regions of the country. We present the construction of the models in detail. Both of them use Markov transition probabilities according to the states of the highway sections and produce a special structure, large-scale, linear programming problem. Some numerical results are presented on the data from Hungarian highways.This work was supported by the National Research Fund, Grant No. 816 and the Ministry of Transport and Telecommunication.     

Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting

Traditional forecasting models are not very effective in most financial time series. To address the problem, this study proposes a novel system for financial modeling and forecasting. In the first stage, wavelet analysis transforms the input space of raw data to a time-scale feature space suitable for financial modeling and forecasting. A spectral clustering algorithm is then used to partition the feature space into several disjointed regions according to their time series dynamics. In the second stage, multiple kernel partial least square regressors ideally suited to each partitioned region are constructed for final forecasting. The proposed model outperforms neural networks, SVMs, and traditional GARCH models, significantly reducing root-mean-squared forecasting errors.     

Infrastructure development for conversion to environmentally friendly fuel

An important concern for any nation wishing to convert to alternate, environmentally friendly energy sources is the development of appropriate fuel distribution infrastructure. We address the problem of optimally locating gas station facilities for developing nations, like India, which are in the process of converting from leaded to unleaded fuel. Importantly, a similar approach may be used in developed countries, which are in the process of converting to automobiles using hydrogen or electrical energy. An integer-programming model with the objective of balancing the perspectives of coverage and cost is presented for this facility location problem. Given the existing network of roads, the model considers the traveling population, the location of existing facilities and the cost of, either converting these facilities to carry unleaded fuel, or of installing new facilities in an attempt to minimize cost and simultaneously maximize coverage of population. We develop a heuristic solution procedure for this problem. The methodology is applied to data sets obtained from Current et al. [J.R. Current, C.S. ReVelle, J.L. Cohon, Decision Sciences 19 (1988) 490] representing the Ohio state limited access highway network, and to the Indian national highway network. Additionally, extensive simulations are carried out in order to examine where our approach compares with the maximum weighted spanning tree approach. This work extends the Maximum Covering/Shortest Path problem (MCSPP) formulated by Current et al. [J.R. Current, C.S. ReVelle, J.L. Cohon, European Journal of Operational Research 21 (1985) 189] to accommodate multiple sources and destinations.     

Forecasting seasonal time series data: a Bayesian model averaging approach

A flexible Bayesian periodic autoregressive model is used for the prediction of quarterly and monthly time series data. As the unknown autoregressive lag order, the occurrence of structural breaks and their respective break dates are common sources of uncertainty these are treated as random quantities within the Bayesian framework. Since no analytical expressions for the corresponding marginal posterior predictive distributions exist a Markov Chain Monte Carlo approach based on data augmentation is proposed. Its performance is demonstrated in Monte Carlo experiments. Instead of resorting to a model selection approach by choosing a particular candidate model for prediction, a forecasting approach based on Bayesian model averaging is used in order to account for model uncertainty and to improve forecasting accuracy. For model diagnosis a Bayesian sign test is introduced to compare the predictive accuracy of different forecasting models in terms of statistical significance. In an empirical application, using monthly unemployment rates of Germany, the performance of the model averaging prediction approach is compared to those of model selected Bayesian and classical (non)periodic time series models.     

Short-term forecasting of traffic delays in highway construction zones using on-line approximators

Highway construction zones are often the cause of traffic delays. This is a natural consequence of the high congestion and nonuniform traffic flow conditions in construction zones. Most of the current algorithms for computing traffic delays are accurate for low density traffic conditions, and address the estimation of current travel time only. This paper presents a method for short-term forecasting of traffic delays in highway construction zones using data from presence detectors. The method is based on a modular approach wherein data from adjacent detectors is processed for estimating the travel time between the two detectors. The travel time estimates are then considered as time-series data, and the problem of short-term forecasting of traffic delay is formulated as a time-series evolution problem. A generic structure referred to as an on-line approximator is used for the prediction of travel time based on current and past travel time estimates. Simulation examples are used to illustrate the traffic delay forecasting algorithm.