多精英采样与个体差分学习的分布估计算法

提出了基于多精英采样和差分搜索的分布估计算法EDA-M/D (Estimation distribution algorithm based on multiple elites sampling and individuals differential search)。EDA-M/D利用多精英个体独立采样生成子代来提升算法全局搜索能力,利用精英群体分布的σ²约束采样半径,实现种群从全局搜索逐步过度到局部搜索。当精英群体停滞时,劣势个体借助精英群体的μ和种群历史最优解进行差分搜索,帮助种群跳出局部最优解。通过多精英采样与差分搜索的自适应协同实现种群宏观信息与个体微观信息的有机融合。实验结果表明EDA-M/D在稳定性和搜索能力方面均表现出明显的优势。     

Optimal sampling frequency for volatility forecast models for the Indian stock markets

This paper evaluates the performance of conditional variance models using high‐frequency data of the National Stock Index (S&P CNX NIFTY) and attempts to determine the optimal sampling frequency for the best daily volatility forecast. A linear combination of the realized volatilities calculated at two different frequencies is used as benchmark to evaluate the volatility forecasting ability of the conditional variance models (GARCH (1, 1)) at different sampling frequencies. From the analysis, it is found that sampling at 30 minutes gives the best forecast for daily volatility. The forecasting ability of these models is deteriorated, however, by the non‐normal property of mean adjusted returns, which is an assumption in conditional variance models. Nevertheless, the optimum frequency remained the same even in the case of different models (EGARCH and PARCH) and different error distribution (generalized error distribution, GED) where the error is reduced to a certain extent by incorporating the asymmetric effect on volatility. Our analysis also suggests that GARCH models with GED innovations or EGRACH and PARCH models would give better estimates of volatility with lower forecast error estimates. Copyright © 2008 John Wiley & Sons, Ltd.     

基于Kriging模型与重要性抽样的可靠性灵敏度分析

针对因无法获得功能函数的梯度信息而不能使用解析方法的情形,提出了进行可靠性灵敏度分析的高效的仿真方法,首先基于Kriging模型和重要性抽样去计算失效概率,然后通过记分函数(score function)方法求出失效概率对各个参数的偏导数。在计算失效概率时采用反问题(inversion problems)中的不确定性逐步减少(stepwise uncertainty reduction)准则来更新功能函数的Kriging模型,继而在重要性抽样的框架下将失效概率表示成一个"增大"的失效概率与修正项的乘积;而记分函数方法只是对前面抽样方法的一个简单后处理,不需要计算额外的功能函数值.对所提方法使用算例验证表明:当功能函数为昂贵的计算模型或对系统(非单个构件)进行灵敏度分析时,该方法具有较高的计算效率和精度。     

Dynamic Latent Class Model Averaging for Online Prediction

We consider the problem of online prediction when it is uncertain what the best prediction model to use is. We develop a method called dynamic latent class model averaging, which combines a state‐space model for the parameters of each of the candidate models of the system with a Markov chain model for the best model. We propose a polychotomous regression model for the transition weights to assume that the probability of a change in time depends on the past through the values of the most recent time periods and spatial correlation among the regions. The evolution of the parameters in each submodel is defined by exponential forgetting. This structure allows the ‘correct’ model to vary over both time and regions. In contrast to existing methods, the proposed model naturally incorporates clustering and prediction analysis in a single unified framework. We develop an efficient Gibbs algorithm for computation, and we demonstrate the value of our framework on simulated experiments and on a real‐world problem: forecasting IBM's corporate revenue. Copyright © 2014 John Wiley & Sons, Ltd.     

Non-invasive genetic study and population monitoring of the brown bear (Ursus arctos) (Mammalia: Ursidae) in Kastoria region – Greece

The brown bear (Ursus arctos) in Greece is considered endangered but little is known about the genetic status and the exact size of local populations. Non-invasive genetic sampling was used in this study to investigate the genetic diversity and genetic structure of the brown bear population in the Kastoria region (northwest Macedonia, Greece) and to estimate its population size. Estimation of demographic parameters was based on innovative, well-evaluated methods that can provide estimates from a single sampling session. DNA was extracted from hairs, scat and blood samples and subsequent amplification of 10 microsatellite loci allowed the identification of a minimum number of 75 living bears in the study area while the mark–recapture-based analysis resulted in a point estimation of 219 individuals. Relatively high diversity values, lack of heterozygosity deficiency as well as estimated effective population size, support the Kastoria bear population having good conservation status.     

尾矿坝振动实验数据样本容量的优化研究

针对铀尾矿坝振动实验逸出氡累积浓度数据样本容量设计的主观性,本文基于偏向角绝对值均值递减还原逸出氡累积浓度曲线,采用随机起点等距抽样方法,得到不同样本容量的数据子集;通过曲线拟合和样本决定系数度量不同子集对还原实验数据集的代表能力。在保障实验研究精度条件下,得到使实验成本较优的最小样本容量,实验结果表明在样本决定系数大于0.98时,优化后样本容量为实验数据样本容量的10%以下。     

基于非等量采样的伪码同步技术研究

针对数字通信系统中伪随机噪声(pseudo-random noise, PN)码同步技术的同步精度受限于采样间隔的问题,结合可变群时延的高精度同步思想,利用采样点位置偏移量的变化,推导出了非等量采样(non-commensurate sampling, NCS)下的PN码相关函数解析表达式,提出了基于精度因子的NCS率选取准则,通过精度因子可快速判断出NCS后PN码的同步精度下限。在此基础上,分析了码序列周期、接收机前端滤波器等附加因素对PN码同步精度的影响。对NCS率选取准则的研究,为优化数字通信系统设计提供了理论指导,达到以低硬件消耗实现高精度同步的目的。     

A New Adaptive Formant Vocoder

A new type of vocoder system based upon formant analysis is presented in this paper. The LMS adaptive algorithm is used for tracking formants of speech signals. The results of computer simulation show that the new vocoder has better synthesized speech quality.     

论复杂适应系统中利益群体博弈的形成过程

复杂适应系统是由多个利益群体所形成的共同利益的集合体,它是协同竞争效应产生的载体。单个利益群体之所以愿意加入博弈的复杂适应系统,是因为它可以借助系统中的协同竞争来获得单靠自身力量所无法获得的价值提升。然而这个过程也正是本文所研究的复杂适应系统中利益群体博弈的形成过程。     

ADAPTIVE STOCHASTIC APPROXIMATION FOR MEASUREMENT WITH CORRELATED NOISE

A multidimensional adaptive stochastic approximation procedure trun-cated at randomly varying bounds is studied for the regression function observedwith correlated noise.The algorithm is proved to be asymptotically efficient.Theresult of the paper generalizes the one shown in[1]for the observation noise be-ing i.i.d.random vectors and under other restrictive conditions imposed on theregression function.