Simultaneous decomposition of the variance by sources and subgroups – new insights

There are two types of decompositions: of linear combinations of random variables into contributions of individual variables (sources) and associations between them, and of populations into contributions of their subpopulations. Simultaneous treatment of the two types is called for, which takes into account the correlations between sources within subpopulations and between subpopulation means. The expected values of the subcomponents are derived and their sensitivity to correlations among sources within groups and among source group means is conducted. An example is provided, in which the correlations contribute 20--25% to total variability. This additional information is hidden when decompositions are not simultaneous.     

Sparse Bayesian variable selection in multinomial probit regression model with application to high-dimensional data classification

Here we consider a multinomial probit regression model where the number of variables substantially exceeds the sample size and only a subset of the available variables is associated with the response. Thus selecting a small number of relevant variables for classification has received a great deal of attention. Generally when the number of variables is substantial, sparsity-enforcing priors for the regression coefficients are called for on grounds of predictive generalization and computational ease. In this paper, we propose a sparse Bayesian variable selection method in multinomial probit regression model for multi-class classification. The performance of our proposed method is demonstrated with one simulated data and three well-known gene expression profiling data: breast cancer data, leukemia data, and small round blue-cell tumors. The results show that compared with other methods, our method is able to select the relevant variables and can obtain competitive classification accuracy with a small subset of relevant genes.     

I'm Not a Bad Mother — Stories of Mothering‐on‐the‐Edge

This article is located in the field of custody and day‐to‐day care, with a focus on experiences of mothering where mothers no longer have everyday care of their children. At the time of writing, tens of thousands of mothers live separately from their children in Aotearoa, New Zealand. However, there is a general perception of rarity associated with non‐resident mothers, partly due to their reluctance to speak of loss of custody. In this article, we develop a theoretical discussion around the inside/outside relationships between non‐resident mothers and their children that informs therapeutic practice. We argue that mothers who live apart from their children are situated at the edge or periphery of their children's day‐to‐day lives and experience this as outside the lives of their children. Nevertheless, mothers' stories speak of qualities of (m)othering that have been born of and nurtured by longstanding and long‐term commitment to their children.     

股票定价模式及股价分解测度方法研究——基于输入输出转换及相对比较的视角

通过超额收益统计检验研究市场有效性和利用均值-方差投资组合理论间接研究金融资产定价的经典方法与范式正面临尖锐的质疑和挑战.本文从输入输出转换和相对比较的视角对股票定价机制进行了分析,提出了一个涵盖四类定价因素,综合反映股票内在价值和相对价值的定价模式,给出了各组成部分的测度模型.最后,利用9家多重上市公司股票论证了本文提出方法的有效性.     

中小板ETF的价格发现能力研究

使用日内5分钟交易高频数据,通过误差修正模型和方差分解等技术研究中小板ETF与其标的指数间的价格发现,进而探讨信息传递过程。实证结果显示：中小板ETF价格与中小板P指数间存在协整关系,达到了长期均衡;价格发现能力上,中小板P指数领先中小板ETF;中小板P指数受到新信息影响所产生的冲击大于中小板ETF价格所产生的冲击,中小板P指数对预测误差方差的解释能力强于中小板ETF价格,中小板P指数为信息传递的领先指标。我国ETF市场的有效性有待提升。     

交易所与银行间债券市场交易机制效率研究

通过构建交易价格分解模型,将交易机制效率进行量化,构造度量交易机制效率综合性指标,该指标充分考虑不同市场交易主体和流动性差异;剔除价差中的逆向选择部分,提取成交价格与有效价格的真实偏离,同时将价格波动归结为由债券新息引起的波动和由交易机制摩擦引起的波动两部分.选取在交易所和银行间市场同时交易的跨市国债作为研究对象,运用逐笔成交高频数据计算交易机制效率综合性指标的平均值和标准差,对两个市场交易机制效率进行对比研究.实证结果表明,交易所债市竞价交易机制价格误差更小,交易机制效率更高;银行间债市较大的报价价差源于做市商的逆向选择风险防范,而做市商机制的真实交易成本与交易所竞价机制相差较小,这种交易成本虽然不利于频繁买卖的现券交易,但对于大宗交易者来说可以忽略.     

Decomposition of K m,n into 4-cycles and 2t-cycles

In this paper we show the sufficient conditions for the decomposition of the complete bipartite graphs K _2m,2n and K _2n+1,2n+1−F into cycles of two different lengths 4 and 2t, t>2, where F is a 1-factor of K _2n+1,2n+1. After that we prove that the results are true for t=5 and 6. Dedicated to Frank K. Hwang on the occasion of his 65th birthday. An erratum to this article can be found at     

Clustering and combinatorial optimization in recursive supervised learning

The use of combinations of weak learners to learn a dataset has been shown to be better than the use of a single strong learner. In fact, the idea is so successful that boosting, an algorithm combining several weak learners for supervised learning, has been considered to be the best off the shelf classifier. However, some problems still exist, including determining the optimal number of weak learners and the over fitting of data. In an earlier work, we developed the RPHP algorithm which solves both these problems by using a combination of global search, weak learning and pattern distribution. In this chapter, we revise the global search component by replacing it with a cluster based combinatorial optimization. Patterns are clustered according to the output space of the problem, i.e., natural clusters are formed based on patterns belonging to each class. A combinatorial optimization problem is therefore created, which is solved using evolutionary algorithms. The evolutionary algorithms identify the “easy” and the “difficult” clusters in the system. The removal of the easy patterns then gives way to the focused learning of the more complicated patterns. The problem therefore becomes recursively simpler. Over fitting is overcome by using a set of validation patterns along with a pattern distributor. An algorithm is also proposed to use the pattern distributor to determine the optimal number of recursions and hence the optimal number of weak learners for the problem. Empirical studies show generally good performance when compared to other state of the art methods.     

Improved Approximation for Breakpoint Graph Decomposition and Sorting by Reversals

Sorting by Reversals (SBR) is one of the most widely studied models of genome rearrangements in computational molecular biology. At present, is the best known approximation ratio achievable in polynomial time for SBR. A very closely related problem, called Breakpoint Graph Decomposition (BGD), calls for a largest collection of edge disjoint cycles in a suitably-defined graph. It has been shown that for almost all instances SBR is equivalent to BGD, in the sense that any solution of the latter corresponds to a solution of the former having the same value. In this paper, we show how to improve the approximation ratio achievable in polynomial time for BGD, from the previously known to for any > 0. Combined with the results in (Caprara, Journal of Combinatorial Optimization, vol. 3, pp. 149–182, 1999b), this yields the same approximation guarantee for n! – O((n – 5)!) out of the n! instances of SBR on permutations with n elements. Our result uses the best known approximation algorithms for Stable Set on graphs with maximum degree 4 as well as for Set Packing where the maximum size of a set is 6. Any improvement in the ratio achieved by these approximation algorithms will yield an automatic improvement of our result.