A Closer Look at the Relation between GARCH and Stochastic Autoregressive Volatility

We show that, for three common SARV models, fitting a minimummean square linear filter is equivalent to fitting a GARCH model.This suggests that GARCH models may be useful for filtering,forecasting, and parameter estimation in stochastic volatilitysettings. To investigate, we use simulations to evaluate howthe three SARV models and their associated GARCH filters performunder controlled conditions and then we use daily currency andequity index returns to evaluate how the models perform in arisk management application. Although the GARCH models produceless precise forecasts than the SARV models in the simulations,it is not clear that the performance differences are large enoughto be economically meaningful. Consistent with this view, wefind that the GARCH and SARV models perform comparably in testsof conditional value-at-risk estimates using the actual data.     

Pricing and hedging power options

In this paper we study the pricing and hedging of options whose payoff is a polynomial function of the underlying price at expiration; so-called ‘power options’. Working in the well-known Black and Scholes (1973) framework we derive closed-form formulas for the prices of general power calls and puts. Parabola options are studied as a special case. Power options can be hedged by statically combining ordinary options in such a way that their payoffs form a piecewise linear function which approximates the power option's payoff. Traditional delta hedging may subsequently be used to reduce any residual risk.     

基于成本优化的制造业生产物流装配线排序问题分析

为了降低制造业的生产成本，提高收益，在分析生产物流配线闲置和超载成本的基础上，采用粒子群算法对其进行优化，并利用模糊自适应改进惯性权值，通过实证研究表明该方法实用，可操作性强，可以有效降低制造业生产物流成本，具有广泛的应用价值。     

Pareto efficient allocation of an in-motion wireless charging infrastructure for electric vehicles in a multipath network

Electric vehicles (EV) use an eco-friendly technology that limits the greenhouse gas emissions of the transport sector, but the limited battery capacity and the density of the battery are the major barriers to the widespread adoption of EV. To mitigate this, a good method seems to be the innovative wireless charging technology called ‘On-Line EV (OLEV)’, which is a contactless electric power transfer technology. This EV technology has the potential to charge the vehicle’s battery dynamically while the vehicle is in motion. This system helps to reduce not only the size of the battery but also its cost, and it also contributes to extending the driving range before the EV has to stop. The high cost of this technology requires an optimal location of the infrastructure along the route. For this reason, the objective of this paper is to study the problem of the location of the wireless charging infrastructure in a transport network composed of multiple routes between the origin and the destination. To find a strategic solution to this problem, we first and foremost propose a nonlinear integer programming solution to reach a compromise between the cost of the battery, which is related to its capacity, and the cost of installing the power transmitters, while maintaining the quality of the vehicle’s routing. Second, we adapt the multi-objective particle swarm optimization (MPSO) approach to our problem, as the particles were robust in solving nonlinear optimization problems. Since we have a multi-objective problem with two binary variables, we combine the binary and discrete versions of the particle swarm optimization approach with the multi-objective one. The port of Le Havre is presented as a case study to illustrate the proposed methodology. The results are analyzed and discussed in order to point out the efficiency of our resolution method.     

基于粒子群算法下的交叉口信号配时优化

为了合理地优化交叉口信号配时,通过分析交叉口的评价指标,建立了以车辆的平均延误、停车次数最小、交叉口整体通行能力最大、各相位有效绿灯时间和交叉口周期时长作为约束条件的数学模型。并通过改进前人研究基础上的动态加权系数,将多目标的非线性优化问题转化为单一目标的非线性规划问题,为了得出更科学稳定的解,提出了改善粒子群算法系统稳定性的2种方法,并将其与粒子群算法结合起来。然后以Matlab为求解工具,结合临邑市某一交叉口实例进行求解分析。最后的结果表明,在使用改进后的粒子群算法进行优化后交叉口通行能力较之现状提升了9%,延误下降了28%,停车次数下降了9%,且各项优化结果均优于Webster,改进后的算法在程序中运行300代,到216代才开始收敛,而未改进的算法稳定性较差,优化结果和收敛曲线则随着实验次数的变化而变化,最后的结论证明了该算法和模型的可靠性。     

基于粒子群算法的证券组合投资模型的研究

从证券组合模型的概念,给出证券组合多目标决策模型。并用模糊优选法将证券组合多目标优化转化为单目标优化。同时,重点描述了粒子群算法,并采用自适应变异的粒子群算法对证券组合投资模型进行求解,最后编程实现证券组合模型的最优解,试验结果表明此方法取得了较好的效果。     

使用粒子群算法解决期权定价模型参数校准问题——以heston模型为例

期权定价模型的参数校准问题是一个常见的难题,以heston 模型为例,定价时需要估计6 个参数,参数估计问题实质上是高维非线性规划问题,由于估参函数的性质不好,一般的估参方法常常失效。使用粒子群（PSO）智能算法可以改善该模型的参数校准问题,因为粒子群算法具有内在随机性,因此参数估计中的局部极小值问题可以被较好地解决。使用2017 年12 月20 日的香港恒生指数期权作为估计样本,并对2017 年12 月25 日的期权进行样本外预测,数值结果表明使用heston 模型对期权进行定价并配合粒子群算法估计参数具有良好的定价效果。     

动态助词“过”的语法化历程

动态助词"过"一般分为"过1"和其中"过2","过1"表示动作的完结,"过2"表示曾然。其语法化历程为:一般动词"过"→趋向动词"过"→动态助词"过1"→动态助词"过2"。动态助词"过1"和"过2"的演变又可分别区分为两个阶段,即演变发生和演变扩散。就形成机制而言,"过1"的产生主要是受隐喻、类推以及重新分析的作用,趋向动词"过"的形成是"过1"产生的关键;而"过2"是"过1"吸收了助词"来"及其经常出现的句法环境的语法意义而产生的,可以说是"过1"在特殊语境下的产物。     

基于分辨力增益的综合识别传感器管理方法

为提高目标综合识别的效能,需对综合识别系统中多种传感器资源进行科学管 理。在分析综合识别中目标优先级和传感器使用约束条件的基础上,建立了利用分辨力增益 作为传感器资源管理优化准则的目标函数,提出了利用传感器混淆矩阵的预测分辨力增益计 算方法,并将粒子群优化算法引入传感器-目标分配NP-hard问题的求解中。仿真结果表明该 方法合理高效。     

基于改进粒子群算法的无人机航迹规划

针对当前基本粒子群算法无人机航迹规划在后期收敛速度比较慢、效率不高、易陷入局部最优等问题,提出一种改进粒子群算法。首先,在迭代前期和后期分段设置惯性权值的调整,实现粒子惯性和寻优行为的平衡;其次,设置一个定值与相邻2次适应度函数最优值比较策略,防止陷入局部最优;最后,引入遗传算法的交叉、变异机制,得出更优的结果。并通过仿真验证了改进粒子群算法在三维空间航迹规划的有效性和可行性。结果表明,与其他航迹规划算法相比,新算法具有路径长度更短、耗时更少、路径更平滑等优点,加快了收敛速度,提高了航迹规划效率和稳定性。因此,改进算法的航迹规划可得到满足约束关系的最优航迹,对实现自主飞行有重要的参考价值。