Optimal determination of the parameters of some biased estimators using genetic algorithm

In this paper, some new algorithms for estimating the biasing parameters of the ridge, Liu and two-parameter estimators are introduced with the help of genetic algorithm (GA). The proposed algorithms are based on minimizing some statistical measures such as mean square error (MSE), mean absolute error (MAE) and mean absolute prediction error (MAPE). At the same time, the new algorithms allow one to keep the condition number and variance inflation factors to be less than or equal to ten by means of the GA. A numerical example is presented to show the utility of the new algorithms. In addition, an extensive Monte Carlo experiment is conducted. The numerical findings prove that the proposed algorithms enable to eliminate the problem of multicollinearity and minimize the MSE, MAE and MAPE.     

索赔次数的开放式混合泊松分布研究

本文建立了索赔次数的多风险类别混合泊松模型。首先，考虑索赔次数的零膨胀、厚尾性和异质性等特征，建立风险类别待定的开放式混合泊松模型，开放式结构使该模型对实际数据的多样特征和风险类别具有良好的自适应性；其次，定义混合权重参数的iSCAD惩罚函数，实现对权重参数的筛选；最后，借助EM算法求得模型参数，实现对各风险类别下索赔次数的估计。借助iSCAD惩罚函数，给出最优混合数，避免传统混合模型中主观选择的弊端，克服传统混合模型中结构复杂、参数估计没有显式表达式、估计结果不便于解释等问题。基于三组风险特征多样数据的实证分析，本文发现该模型可以显著改进现有模型的拟合效果。     

高维大数据基因网络中的社区发现——以NC方法为例

从大量基因中识别出致病基因是大数据下的一个十分重要的高维统计问题。基因间网络结构的存在使得对于致病基因的识别已从单个基因识别扩展到基因模块识别。从基因网络中挖掘出基因模块就是所谓的社区发现（或节点聚类）问题。绝大多数社区发现方法仅利用网络结构信息，而忽略节点本身的信息。Newman和Clauset于2016年提出了一个将二者有机结合的基于统计推断的社区发现方法（简称为NC方法）。本文以NC方法为案例，介绍统计方法在实际基因网络中的应用和取得的成果，并从统计学角度提出了改进措施。通过对NC方法的分析可以看出对于以基因网络为代表的非结构化数据，统计思想和原理在数据分析中仍然处于核心地位。而相应的统计方法则需要针对数据的特点及关心的问题进行相应的调整和优化。     

An upper bound on the minimal total cost of the transportation problem with varying demands and supplies

In general cases, to find the exact upper bound on the minimal total cost of the transportation problem with varying demands and supplies is an NP-hard problem. In literature, there are only two approaches with several shortcomings to solve the problem. In this paper, the problem is formulated as a bi-level programming model, and proven to be solvable in a polynomial time if the sum of the lower bounds for all the supplies is no less than the sum of the upper bounds for all the demands; and a heuristic algorithm named TPVDS-A based on genetic algorithm is developed as an efficient and robust solution method of the model. Computational experiments on benchmark and new randomly generated instances show that the TPVDS-A algorithm outperforms the two existing approaches.     

网络社区发现算法在流动表建模中的设计与应用

代际流动表可以统计子代与其父代社会地位配对数据的交互频数，反映了社会资源占有的优劣势在父子两代人之间的比较。对财富、阶级、特权等社会基本特征演变的实证考察，均依赖于代际流动表的量化分析。对数线性模型是流动表建模分析的基本工具，通过对列联表单元格频数进行拟合，可以识别流动表行分类与列分类之间的强弱交互效应，刻画父子社会地位间的交互结构。本文利用复杂网络社区发现算法分析父子社会地位的关联结构，针对简约对数线性模型拟合精度不够的问题，提出一种新的建模思路：利用社区发现算法对简约对数线性模型的残差列联表进行关联关系挖掘，将发现的社区效应作为附加参数约束引入原对数线性模型，以改善数据的拟合情况。由于该方法只在原简约对数线性模型中增加了一个参数约束，因此仍可以保证建模结果的简洁性及理论意义，同时社区效应补充了原对数线性模型对经验数据结构的解读。论文用此方法对来源于中国综合社会调查数据的经验代际职业流动表进行建模分析，较好地解释了子代职业阶层与父代职业阶层间的关联模式。     

Multi-objective optimization in partitioning the healthcare system of Parana State in Brazil

Motivated by a proposal of the local authority for improving the existing healthcare system in the Parana State in Brazil, this article presents an optimization-based model for developing a better system for patients by aggregating various health services offered in the municipalities of Parana into some microregions. The problem is formulated as a multi-objective partitioning of the nodes of an undirected graph (or network) with the municipalities as the nodes and the roads connecting them as the edges of the graph. Maximizing the population homogeneity in the microregions, maximizing the variety of medical procedures offered in the microregions, and minimizing the inter-microregion distances to be traveled by patients are considered as three objective functions of the problem. An integer-coded multi-objective genetic algorithm is adopted as the optimization tool, which yields a significant improvement to the existing healthcare system map of the Parana State. The results obtained may have a strong impact on the healthcare system management in Parana. The model proposed here could be a useful tool to aid the decision-making in health management, as well as for better organization of any healthcare system, including those of other Brazilian States.     

Nonlinear and non-gaussian state estimation: A quasi-optimal estimator

The rejection sampling filter and smoother, proposed by Tanizaki (1996, 1999), Tanizaki and Mariano (1998) and Hiirzeler and Kiinsch (1998), take a lot of time computationally. The Markov chain Monte Carlo smoother, developed by Carlin, Poison and StofFer (1992), Carter and Kohn (1994, 1996) and Geweke and Tanizaki (1999a, 1999b), does not show a good performance depending on noniinearity and nonnormality of the system in the sense of the root mean square error criterion, which reason comes from slow convergence of the Gibbs sampler. Taking into account these problems, we propose the nonlinear and non-Gaussian filter and smoother which have much less computational burden and give us relatively better state estimates, although the proposed estimator does not yield the optimal state estimates in the sense of the minimum mean square error. The proposed filter and smoother are called the quasi-optimal filter and quasi-optimal smoother in this paper. Finally, through some Monte Carlo studies, the quasi-optimal filter and smoother are compared with the rejection sampling procedure and the Markov chain Monte Carlo procedure.     

Impact of investing in quality improvement on (Q,r, L) model involving the imperfect production process

This paper investigates the impact of quality improvement on the modified lot size reorder point models involving variable lead time and partial backorders. The formulated models include the imperfect production process and an investing option of improving the process quality. The objective is simultaneously optimizing the lot size, reorder point, process quality level and lead time. We first assume that the lead time demand follows a normal distribution, then relax this assumption to consider the distribution-free case where only the mean and standard deviation of lead time demand are known. An algorithm procedure of finding the optimal solution is developed, and two numerical examples are given to illustrate the results.     

基于自组织算法的改进型GAANN预测模型

在传统的基于GA算法人工神经网络的基础上作了改进,将训练集分为两部分,在前一训练集训练后获得的网络基础上使用后一训练集进行进一步的训练获得更为优化的网络结构。针对复杂系统建模输入节点难以确定的问题,提出将其与自组织理论相结合,首先使用GMDH方法获得神经网络的初始化节点,然后使用训练好的神经网络模型进行预测。最后,将由此建立的预测模型应用于国家粮食产量预测,取得了令人满意的效果。     

Quadratic estimation in mixed linear models with two variance components

Explicit expressions for Bayes invariant quadratic estimates, biased and unbiased, are presented and proved to cover the entire class of admissible estimates in the considered classes. An unbalanced genetic model is studied for demonstration.