基于Hawkes过程中美股市大幅波动互激效应的研究

近年来，由于中美经济联系日趋紧密，中美股票市场大幅波动的互激效应明显增强。本文考虑中美股市时差和法定节假日差异等因素，运用标值Hawkes过程对2006-2017年CSI300和S&P500大幅波动收益率数据进行建模，结果表明：（1）中美股市大幅波动互激效应存在不对称性，美股市场大幅波动对中国股市的互激效应更强；（2）中美股市大幅波动的幅度对互激效应不存在显著影响；（3）中美股票市场对于大幅波动互激效应的消化速度存在差异，中国股票市场消化美股大幅波动互激效应的速度较快。本研究对金融市场监管者和投资者均有一定意义。     

Precise large deviations for the difference of two sums of END random variables with heavy tails

Assume that there are two types of insurance contracts in an insurance company, and the ith related claims are denoted by {X_ij, j ? 1}, i = 1, 2. In this article, the asymptotic behaviors of precise large deviations for non random difference ∑^n₁(t)_{j = 1}X_1j ? ∑^n₂(t)_{j = 1}X_2j and random difference ∑^N₁(t)_{j = 1}X_1j ? ∑^N₂(t)_{j = 1}X_2j are investigated, and under several assumptions, some corresponding asymptotic formulas are obtained.     

Complete convergence and the strong laws of large numbers for pairwise NQD random variables

The authors study the strong convergence for sequences of pairwise negatively quadrant dependent (NQD) random variables under some wide conditions, and present some new theorems on the complete convergence and the strong laws of large numbers. The obtained results extend and improve some theorems in existing literature.     

Strong law of large numbers and complete convergence for non-identically distributed WOD random variables

In this paper, we establish the strong law of large numbers and complete convergence for non-identically distributed WOD random variables. We derive some new inequalities of Fuk–Nagaev type for the sums of non-identically distributed WD random variables. All these results further extend and refine previous ones.     

Elementary renewal theorems for widely dependent random variables with applications to precise large deviations

Abstract

On the basis of Wang and Cheng (J. Math. Anal. Appl. 384 (2011) 597–606), this paper further investigates elementary renewal theorems for counting processes generated by random walks with widely orthant dependent increments. The obtained results improve the corresponding ones of the above-mentioned paper mainly in the sense of weakening the moment conditions on the positive parts of the increments. Meanwhile, a revised version of strong law of large numbers for random walks with widely orthant dependent increments is established, which improves Theorem 1.4 of Wang and Cheng (2011 Wang, Y., and D. Cheng. 2011. Basic renewal theorems for a random walk with widely dependent increments and their applications. Journal of Mathematical Analysis and Applications 384 (2):597–606. doi:10.1016/j.jmaa.2011.06.010.[Crossref], [Web of Science ®] , [Google Scholar]) by enlarging the regions of dominating coefficients. Finally, by using the above results, some precise large deviation results for a nonstandard renewal risk model are established, in which the innovations are widely orthant dependent random variables with common heavy tails, and the inter-arrival times are also widely orthant dependent.     

《孟子》释读辨误(四则)

传统经典须细读，在细读的基础上领会真义。历代注家对《孟子》的注疏多有误读之处，如"攘"，本义为"取自来之物"，经"多少同盗"，演变为"攘"即"盗"；"巨室"初谓国君住的大房子，这里当喻指国君，但注家多强解为指大臣；舜在"完廪""浚井"的过程中是如何脱难的，本文没有说，注家依据其他材料做了合理化解读，未必符合本义。"坏宫室以为汙池"，如从人类社会发展史的角度来释读，实际上是要维持以渔猎为生的生存方式。致误的根源有"非黑即白"的思维方式、前后矛盾、对古代"不雅驯"的文字进行改造等。     

基于ODR-ADASYN-SVM的极端金融风险预警研究

针对合成少数类过采样(synthetic minority over-sampling technique,SMOTE)方法在提升支持向量机(support vector machine,SVM)的非均衡样本学习能力中出现的过拟合(over fitting),引入自适应合成抽样方法(adaptive synthetic sampling approach,ADASYN)和逐级优化递减欠采样方法(optimization of decreasing reduction,ODR)分别克服SMOTE在生成新样本中的盲目性和在处理对象上的局限性,进而与SVM相结合,构造出改进SVM,即ODR-ADASYN-SVM模型来预测中国极端金融风险;最后运用T检验对各模型预测精度的差异性进行显著性检验以及对各模型的预测稳定性进行评价.实证结果表明,ODR-ADASYN-SVM模型不仅能够显著地提升SVM的非均衡样本学习能力,同时也能够有效地克服SMOTE的过拟合,从而展示出优越的极端金融风险预测性能.     

Information in Tender Offers With a Large Shareholder

We study takeovers of firms whose ownership structure is a mixture of minority block‐holders and small shareholders. We show that the combination of dispersed private information on the side of small shareholders and the presence of a large shareholder can facilitate profitable takeovers. Furthermore, our analysis implies that even if some model of takeovers predicts a profit for the raider, for example, due to private benefits, the profit will be underestimated unless the large shareholder and the dispersion of information among the small shareholders are modeled.     

A strong law of large numbers for independent random variables under non-additive probabilities

Abstract

Under non‐additive probabilities, cluster points of the empirical average have been proved to quasi-surely fall into the interval constructed by either the lower and upper expectations or the lower and upper Choquet expectations. In this paper, based on the initiated notion of independence, we obtain a different Marcinkiewicz-Zygmund type strong law of large numbers. Then the Kolmogorov type strong law of large numbers can be derived from it directly, stating that the closed interval between the lower and upper expectations is the smallest one that covers cluster points of the empirical average quasi-surely.     

发电成本垂直差异电力市场概率发电——基于大用户电量偏好视角

针对碳减排后普遍存在的发电成本垂直差异电力市场及其显著的大用户电量偏好差异,本文提出概率性电量供给以及与之对应的概率发电,并利用价格杠杆调控发电商低碳发电尤其高发电成本发电机会与供给效能不足.先后构建发电成本内、外生选择下的概率发电模型,求解发电商的电量供给均衡并分析概率发电策略的最优性.最后,将发电成本内生选择模型拓展到分解市场需求不确定性的概率发电模型.研究发现:相比基准发电,概率发电可以作为一种有利可图的发电容量过剩的解决方法,即使在发电成本内生选择时也是可行的;当发电商采用“强”成本差异的发电策略时,引入折中的概率发电会让产品线上成本水平相互更接近,且大用户的消费者剩余得到提高;与此相反,在市场中实施“弱”成本差异发电,发电商利用概率发电则增大成本分离并降低大用户的消费者剩余;在不确定性市场需求下,当偏好低发电成本电量的大用户对单位发电成本的估价相对于偏好高发电成本电量的大用户对单位发电成本的估价足够高时,概率发电可以以一种管理不利需求状况工具的身份出现.