Hilbert Spectra and Empirical Mode Decomposition: A Multiscale Event Analysis Method to Detect the Impact of Economic Crises on the European Carbon Market

Exploring the effect of an economic crisis on the carbon market can be propitious to understand the formation mechanisms of carbon pricing, and prompt the healthy development of the carbon market. Through the ensemble empirical mode decomposition (EEMD), a multiscale event analysis approach is proposed for exploring the effect of an economic crisis on the European carbon market. Firstly, we determine the appropriate carbon price data of the estimation and event windows to embody the impact of the interested economic crisis on carbon market. Secondly, we use the EEMD to decompose the carbon price into simple modes. Hilbert spectra are adopted to identify the main mode, which is then used to estimate the strength of an extreme event on the carbon price. Thirdly, we perform a multiscale analysis that the composition of the low-frequency modes and residue is identifying as the main mode to capture the strength of the interested economic crisis on the carbon market, and the high-frequency modes are identifying as the normal market fluctuations with a little short-term effect on the carbon market. Finally, taking the 2007–2009 global financial crisis and 2009–2013 European debt crisis as two cases, the empirical results show that contrasted with the traditional intervention analysis and event analysis with the principle of “one divides into two”, the proposed method can capture the influences of an economic crisis on the carbon market at various timescales in a nonlinear framework.     

考虑产能需求方交期和价格敏感的产能分享平台定价策略

基于双边市场理论,考虑产能分享双边市场具有网络外部性特性,同时考虑产能需求方对加工交期和价格敏感,研究垄断型制造业产能分享平台的定价策略.首先,建立平台和双边用户的两阶段决策模型;然后,通过计算求解探索了注册费和交易费收费模式下的平台均衡利润,并分析了交叉网络外部性等外生变量对各方决策和平台利润的影响.研究发现:注册费模式下的平台均衡利润大于固定交易费模式下的平台均衡利润;两种定价模式下的平台利润与双边用户的网络外部性均正相关,与产能需求方的交期预期偏差均负相关,与产能需求方的产能价格预期均正相关;产能需求方的交期敏感度在实际交期早于或晚于预期交期时对平台利润有不同影响,产能需求方的产能价格敏感度在产能实际价格低于或高于预期价格时对平台利润也有不同影响.     

基于动态服务价的充电站有序充电控制方法

由于电动汽车的不断增长与不均衡发展,社会中的大多充电站存在着部分充电设施闲置与部分充电排队并存的现象。为了解决电动汽车充电困难与充电设施利用率低这一对新矛盾,本文首先调查分析了各种电动车辆的出行习惯和充电特点,然后对社会充电站归纳成城市公共充电站和公交物流车等专用充电站两类,提出了分时电价响应和动态服务价响应两种有序充电方式,并制定以风电、光伏等绿色新能源出力反向激励的动态服务价引导机制,建立了以用户充电费用最低为目标的数学模型。最后通过仿真分析,分时电价响应、动态服务价响应有序充电方式下均比无序充电方式降低负荷波动性,动态服务价响应有序充电方式能够有效提高西部地区绿色电能利用率,大幅减少弃风弃光,为环境保护、能源结构调整起到重要作用。     

A hybrid VMD–BiGRU model for rubber futures time series forecasting

As one of the four major industrial raw materials in the world, natural rubber is closely related to the national economy and people’s livelihood. The analysis of natural rubber price and volatility can give hedging guidance to manufacturers and provide investors with uncertainty and risk information to reduce investment losses. To effectively analyses and forecast the natural rubber’s price and volatility, this paper constructed a hybrid model that integrated the bidirectional gated recurrent unit and variational mode decomposition for short-term prediction of the natural rubber futures on the Shanghai Futures Exchange. In data preprocessing period, time series is decomposed by variational mode decomposition to capture the tendency and mutability information. The bidirectional gated recurrent unit is introduced to return the one-day-ahead prediction of the closing price and 7-day volatility for the natural rubber futures. The experimental results demonstrated that: (a) variational mode decomposition is an effective method for time series analysis, which can capture the information closely related to the market fluctuations; (b) the bidirectional neural network structure can significantly improve the model performance both in terms of fitting performance and the trend prediction; (c) a correspondence was found between the predicted target, i.e., the price and volatility, and the intrinsic modes, which manifested as the impact of the long-term and short-term characteristics on the targets at different time-scales. With a change in the time scale of forecasting targets, it was found that there was some variation in matching degree between the forecasting target and the mode sub-sequences.     

具有网络外部效应的三度价格歧视研究

通过建立两厂商和两子市场的两阶段博弈模型,分析具有网络外部效应的寡头竞争市场厂商实施三度歧视定价的产出、价格和社会福利问题,研究结果表明:厂商通过歧视定价提高了强市场的价格,相应降低了弱市场的价格,虽然总产出没有改变,但却减少了社会总福利.     

基于模糊近似支持向量回归的股价预测研究

股价预测是投资策略形成和风险管理模型发展的基础。为了降低股价变化趋势中的噪声信息和投资者关于两种股价预测误差的不同偏好对股价预测的影响,提出了基于信噪比的模糊近似支持向量回归(FPSVR)的股价预测模型。首先构建信噪比输入变量,然后引入模糊隶属度和双边权重测量方法对支持向量回归(SVR)模型进行改进,最后借助沪深300成份股2008至2019年的股票时间序列日数据,按照股市的波动情况将其分为三个阶段(牛市、熊市、震荡市),并建立三个基准模型进行对比分析。研究结果表明:与三个基准模型相比,所提出的股价预测模型的预测误差最低;与原有的SVR模型相比,FPSVR模型可以更好地对处于牛市和震荡市阶段的股票时间序列进行股价预测。     

人工神经网络在证券价格预测中的应用

证券市场中成功的交易模式是可以模仿及学习的.证券价格走势实质是一种复杂时序函数.人工神经网络是在模仿人脑处理问题过程中发展起来的新型智能信息处理系统,人工神经网络可以通过调节连接权值以任意精度逼近任何连续函数,因此也可以逼近证券价格随时间变换这种函数.文中采用基于BP模型的神经网络,用BP算法和遗传算法来训练网络权值,同时也采用了动量法和学习率自适应调整相结合的策略,对证券市场的价格进行建模和预测,结果表明,此模型具有较好的学习、泛化能力,对股票市场或其他类似的非线性经济系统的走势预测决策具有较好的效果.     

二层分解技术在电价预测中的应用研究

针对电价波动幅度大且预测精度低的问题,提出了二层分解技术与神经网络组成的电价多步预测模型。该模型首先采用集合经验模式分解将原始电价序列分解为一系列分量,变分模态分解将第一层分解产生的最高频率分量进一步分解为一系列模态分量,所有分量采用神经网络模型进行预测,并使用纵横交叉算法对神经网络的参数进行优化,最后叠加所有子序列,得出预测电价值。仿真结果表明,所提出的模型相比其他混合模型具有更好的预测性能,且实用价值高。     

基于非结构数据流行学习的碳价格多尺度组合预测

碳交易价格的有效预测对制定符合国情的碳金融市场政策以及碳金融市场的风险管理都具有重要意义.对此,提出一种基于非结构数据流行学习的碳价格多尺度组合预测方法.首先,利用网络搜索指数提取碳价格相关的非结构化数据,基于等度量映射流行学习对其进行降维;然后,对降维后的非结构化数据、其他影响因素结构化数据、碳交易价格分别进行经验模态分解(Empirical mode decomposition,EMD),得到不同个数的本征模函数(Intrinsic mode function,IMF),并采用Fine-to-coarse方法对IMF进行重构,得到高频序列、低频序列和趋势项;最后,利用自回归积分滑动平均模型(Autoregressive integrated moving average model,ARIMA)、偏最小二乘(Partial least squares,PLS)回归和神经网络对高频数据、低频数据和趋势项进行预测,将3种预测结果进行集成,得到最终预测值.仿真实验结果表明,所提出的方法可以有效利用多源信息,具有较高的预测精度和良好的适用性.     

基于EMD改进的Elman神经网络对股票的短期预测模型

Elman神经网络在短期预测股市收盘价时存在预测趋势良好但准确度较低的问题。在Elman神经网络的思想上提出以经验模态分解EMD为基础的Elman新组合模型。应用EMD将各交易日的收盘价序列分解成不同时间尺度上的本征模函数IMF分量和剩余分量,进而利用偏自相关函数PACF计算每一个分量的滞后期,以确定各分量在Elman神经网络中的输入和输出变量,从而得到各分量的预测值,相加得到最终的预测结果。与EMD单一网络、EMD-Elman模型、BP网络及EMD-BP模型进行实验对比,结果表明:该短期预测模型的预测值均方误差、平均绝对误差和平均绝对百分比误差都得到较大的改善;新组合模型可有效实现对股票收盘价的短期预测,且能降低非平稳性对预测结果的影响。该研究为进一步预测股市的走向提供了有效依据,也为投资者提供了更充分的决策参考。     

Real‐time waiting‐price trading interval in a heterogeneous options market: a Bernoulli distribution

Options pricing remains an open research question that is challenging for both theoreticians and practitioners. Unlike many classical binomial models that assume a “representative agent,” the model suggested herein considers two players who are heterogeneous with respect to their estimations of the distribution of the underlying asset price on expiration day, and with respect to their levels of willingness to make a transaction (eagerness level). A two‐player binomial model is developed to find the real‐time optimal option price in two stages. First, we determine a primary feasible pricing domain. We then find a narrower feasible domain, termed the “waiting‐price trading interval,” meaning the region within which the players may either wait for better offers (due to a change in market conditions or player beliefs), or make an immediate transaction. The suggested model is formulated by a nonlinear optimization problem and the optimal price is shown to be unique. We demonstrate that the counter player's eagerness level has a significant effect on the proposed optimal option price. Using empirical analysis, several known lattice‐based models for option pricing, such as CRR and Tian, are compared with the current model (herein, S‐H) in which the price offered by the model player takes into account the subjective beliefs of the opposing market player. The comparison shows significant advantages to the S‐H model in terms of the expected profit on expiration day.     

基于ATLG混合模型的股票价格预测

股市是金融市场的重要组成部分,对股票价格预测有着重要的意义.同时,深度学习具有强大的数据处理能力,可以解决金融时间序列的复杂性所带来的问题.对此,本文提出一种结合自注意力机制的混合神经网络模型(ATLG).该模型由长短期记忆网络(LSTM)、门控递归单元(GRU)、自注意力机制构建而成,用于对股票价格的预测.实验结果表明:(1)与LSTM、GRU、RNN-LSTM、RNN-GRU等模型相比, ATLG模型的准确率更高;(2)引入自注意力机制使模型更能聚焦于重要时间点的股票特征信息;(3)通过对比,双层神经网络起到的效果更为明显.(4)通过MACD (moving average convergence and divergence)指标进行回测检验,获得了53%的收益,高于同期沪深300的收益.结果证明了该模型在股票价格预测中的有效性和实用性.     

Inventory model for seasonal demand with option to change the market

In real life, when bulk purchase becomes convenient or even mandatory, it is a common practice for distributors to explore an alternative market in order to maximize the revenue earned. In this paper, we consider an inventory model for a product having seasonal demand with two potential markets, say, primary and alternate. The distributor has a single opportunity of procurement prior to multiple demand seasons in the primary and the alternate market. Both the markets have similar demand patterns, with time lag between their demand seasons. The demand is a price and time dependent function with increasing, constant and declining phases within each demand season. The scale parameter of demand rate depends upon the market. In each market, successive seasons are separated by random time. In one replenishment cycle, the distributor has a single option to exit the primary market by transferring the inventory, with or without change in selling price. This option can be exercised at the end of any complete season at the primary market. Our investigations indicate that it will be beneficial for the distributor to shift to the alternate market even at a slightly lower selling price if demand rate in the alternate market is higher. Optimal number of seasons at the primary market before change of price or market is obtained. Optimal policy is obtained for jointly determining the order quantity and price. Concavity of the profit function is discussed. Solution procedure, numerical examples and sensitivity analysis are presented.     

Inventory model for seasonal demand with option to change the market

In real life, when bulk purchase becomes convenient or even mandatory, it is a common practice for distributors to explore an alternative market in order to maximize the revenue earned. In this paper, we consider an inventory model for a product having seasonal demand with two potential markets, say, primary and alternate. The distributor has a single opportunity of procurement prior to multiple demand seasons in the primary and the alternate market. Both the markets have similar demand patterns, with time lag between their demand seasons. The demand is a price and time dependent function with increasing, constant and declining phases within each demand season. The scale parameter of demand rate depends upon the market. In each market, successive seasons are separated by random time. In one replenishment cycle, the distributor has a single option to exit the primary market by transferring the inventory, with or without change in selling price. This option can be exercised at the end of any complete season at the primary market. Our investigations indicate that it will be beneficial for the distributor to shift to the alternate market even at a slightly lower selling price if demand rate in the alternate market is higher. Optimal number of seasons at the primary market before change of price or market is obtained. Optimal policy is obtained for jointly determining the order quantity and price. Concavity of the profit function is discussed. Solution procedure, numerical examples and sensitivity analysis are presented.     

基于深度迁移学习的多尺度股票预测

股票市场不仅是上市公司的重要融资渠道,也是重要的投资市场,股票预测一直受到人们的关注。为了充分利用来自不同股票价格的信息,提高股票的预测效果,提出一种多尺度股票价格预测模型TL-EMD-LSTM-MA（TELM）。TELM模型通过经验模态分解将收盘价分解为多个时间尺度分量,不同时间尺度分量震荡频率不同,反映了不同的周期性信息;根据分量的震荡频率选择不同方法进行预测,高频分量利用深度迁移学习的方法训练堆叠LSTM,低频分量利用移动平均法进行预测;将所有分量的预测值相加作为收盘价的最终预测输出。通过深度迁移学习训练的堆叠LSTM,包含来自不同股票的信息,具备更多行业或市场的知识,能有效降低预测误差。利用移动平均法预测低频分量,更有效捕获股票的总体趋势。对中国A股市场内500支股票以及上证指数、深证成指等指数进行预测,结果表明,与其他模型相比,TELM预测误差最低,拟合优度最高。根据TELM预测的股票收盘价模拟股票交易过程,结果表明TELM投资风险低、收益高。     

聚合模式下网约车市场的定价策略研究

针对聚合模式给网约车市场带来的收益分配和价格竞争问题,引入平台抽成和出行服务商服务差异刻画聚合模式下网约车市场的特征,构建了聚合平台和两服务商的斯坦伯格博弈模型,探讨了在聚合模式下网约车市场的定价均衡。随后引入平台补贴定价策略,分析了补贴策略对市场均衡定价以及聚合平台、出行服务商期望收益的影响。研究表明：聚合模式下出行服务商服务差异是影响平台抽成比例和网约车市场服务定价的重要因素,对聚合平台和出行服务商的期望收益有双向影响;平台对出行者的价格补贴提高了补贴前网约车市场的服务定价,导致平台抽成增加;采用合理的补贴定价策略可以有效提升出行服务商和聚合平台的收益,同时改进社会福利。     

基于分布式神经网络的苹果价格预测方法

针对传统农产品价格预测模型在大数据场景下无法快速准确对苹果市场价格进行预测的问题,提出一种基于分布式神经网络的苹果价格预测方法。首先,研究影响苹果市场价格的相关因素,选取苹果历史价格、替代品历史价格、居民消费水平和原油价格四个特征作为神经网络模型的输入;然后,构建蕴含价格波动规律的分布式神经网络模型,实现对苹果市场价格的短期预测。实验结果显示,基于分布式神经网络的苹果市场价格短期预测模型具有较高的预测精度,平均相对误差仅为0.50%,满足苹果市场价格预测的要求。实验结果表明,分布式神经网络模型能够通过自学习特性揭示出苹果市场价格的波动规律和发展趋势,所提方法能为稳定苹果市场秩序和市场价格宏观调控提供科学依据,有助于降低价格波动带来的危害,帮助果农规避市场风险。     

基于Elman神经网络的股票价格预测研究

为了更好地把握股票价格的波动，应用了在处理序列数据输入输出具有优越性的Elman 递归神经网络建立股市预测模型，并用两支股票进行了检测，检测结果说明人工神经网络应用于中国股票市场的预测是可行和有效的，有着良好的前景。     

资本市场系统结构模型应用研究

根据资产价格与交易者的相互影响机制，提出了资本市场结构模型及其数学原理，用于分析一组市场资产的价格演变过程，并对其拟合能力进行了探讨，随后进行了实证研究．理论研究和实证分析均表明，资本市场结构模型具有较强的整体拟合能力，能较好地对资产价格进行多指标预测．     

Modes decomposition method in fusion with robust random vector functional link network for crude oil price forecasting

This paper proposes a decomposition based method in fusion with the non-iterative approach for crude oil price forecasting. In this approach, the robust random vector functional link network (RVFLN), a non-iterative approach in fusion with the most efficient decomposition technique called variational mode decomposition (VMD) is proposed which is executed with two links — fixed assigned random weights and direct link from input to output, and the iterative learning process is not involved in its functioning which makes it faster in execution as compared to many existing techniques proposed for forecasting. The fusion of VMD and robust RVFLN called VMD-RVFLN is implemented for crude oil price forecasting where the crude oil price series is decomposed using VMD into a linear smoother series by extracting useful information and the decomposed modes pass through the robust RVFLN model which produces the final forecasting values. The analysis performed in the study approves its efficiency and reports improvement in forecasting accuracy and execution time as compared to some of the traditional iterative techniques like BPNN (back propagation neural network), ARIMA (auto-regressive integrated moving average), LSSVR (least squares support vector regression), ANFIS (adaptive neuro-fuzzy inference system), IT2FNN (interval type-2 fuzzy neural network) and RNN (recurrent neural network), etc. However, both ELM and RVFLN without modes decomposition fusion exhibit less execution time at the cost of reduction in prediction accuracy.