基于复杂网络的电动汽车充电桩众筹市场信用风险建模与分析

因其经济优势及环境友好特性，电动汽车（EV）保有量持续攀升。而当前中国充电设施建设明显滞后于电动汽车的扩张速度，成为制约电动车发展潜力的主要瓶颈。由于充电设施具有高正外部性，同时投资回收期较长，难以吸引到传统融资，而基于互联网的众筹作为新型融资方式则受到了广泛关注。但由于缺乏强制性的信息披露制度及相关法律法规，众筹融资过程中的信息不对称问题相较传统融资方式更为严重，由此导致的高违约率也进一步削弱了投资者信心。本文利用复杂网络对电动汽车充电桩众筹融资市场进行建模，刻画了众筹投资者决策过程中的羊群效应，基于信息不对称使用多Agent模型对信用风险的产生、影响及针对性政策效果进行了模拟分析。结果表明，众筹投资者在决策过程中易受到社交网络中负面信息的影响，使得筹资人违约行为的不利影响进一步扩散。而严格的罚款机制可以在不影响融资市场均衡的情况下，使违约行为的恶性循环得到遏制。     

结构型一般EV线性模型参数估计的相合性研究

本文对结构型一般EV线性模型参数的研究,利用矩方法,得到了未知参数的强相合性。     

Complete convergence for weighted sums of mixingale sequences and statistical applications

In this paper, the complete convergence of weighted sums of L^r-mixingale is established, from which the complete convergence of martingale differences is also derived. As statistical applications, non parametric regression model and simpler linear errors-in-variables model with mixingale errors are discussed.     

Complete convergence for weighted sums of i.i.d. random variables with applications in regression estimation and EV model

In this paper, we obtain complete convergence results for Stout type weighted sums of i.i.d. random variables. A strong law for weighted sums of i.i.d. random variables is also obtained. As the applications of the strong law, the strong consistency and rate of the nonparametric regression estimations and the rates of the strong consistency of LS estimators for the unknown parameters of the simple linear errors in variables (EV) model are given.     

Estimation of the linear EV model with censored data

In this paper, a censored linear errors-in-variables model is investigated. The asymptotic normality of the unknown parameter's estimator is obtained. Two empirical log-likelihood ratio statistics for the unknown parameter in the model are suggested. It is proved that the proposed statistics are asymptotically chi-squared under some mild conditions, and hence can be used to construct the confidence regions of the parameter of interest. Finite sample performance of the proposed method is illustrated in a simulation study.     

Asymptotic Normality for the Partially Linear EV Models with Longitudinal Data

In this article, we consider a partially linear EV regression model under longitudinal data. By using a weighted kernel method and modified least-squared method, the estimators of unknown parameter, the unknown function are constructed and the asymptotic normality of the estimators are derived. Simulation studies are conducted to illustrate the finite-sample performance of the proposed method.     

Some Limit Behaviors for Linear EV Model with Replicate Observations

In this article, we establish the asymptotic normality for the LS estimators of the unknown parameters in the linear errors-in-variables (EV) model with replicate observations, which weaken the moment conditions of Zhang and Chen (2002 Zhang, S.G., Chen, X.R. (2002). On asymptotic normality of parameters in linear EV model. Chin. Ann. Math. Ser. B 23(4):495506.[Crossref], [Web of Science ®] , [Google Scholar]). Furthermore, the moderate and large deviations principles for the LS estimators are obtained also.     

纵向部分线性变系数EV模型的估计

纵向数据是随着时间变化对个体进行重复观测而得到的一种相关性数据，广泛出现在诸多科学研究领域。在对个体进行观测时，测量误差不可避免，忽略测量误差往往会导致有偏估计。本文利用二次推断函数方法研究关于纵向数据的参数部分和非参数部分协变量均含有测量误差的部分线性变系数测量误差(errors-in-variables, EV)模型的估计问题。利用B样条逼近模型中的未知系数函数，构造关于回归参数和B样条系数的偏差修正的二次推断函数以处理个体内相关性和测量误差，得到回归参数和变系数的偏差修正的二次推断函数估计，然后证明了估计方法和结果的渐近性质。数值模拟和实例数据分析结果显示本文提出的方法具有一定的实用价值。