Constrained Bayesian Method for Testing the Directional Hypotheses

The paper discusses the generalization of constrained Bayesian method （CBM） for arbitrary loss functions and its application for testing the directional hypotheses. The problem is stated in terms of false and true discovery rates. One more criterion of estimation of directional hypotheses tests quality, the Type III errors rate, is considered. The ratio among discovery rates and the Type III errors rate in CBM is considered. The advantage of CBM in comparison with Bayes and frequentist methods is theoretically proved and demonstrated by an example.

Simple Statistical Tests Used in Positive Accounting Research： a Thing to be Noted

This paper is concerned with the right use of simple hypothesis tests in the area of positive accounting research. Both the usefulness and the limitations of the technique are dealt with in detail.

Impact of the Qinghai-Tibet Railway on Population Genetic Structure of the Toad-Headed Lizard,Phrynocephalus vlangalii

Using data from nine microsatellite DNA loci and a population genetic approach,we evaluate the barrier effect of the Qinghai-Tibet Railway on toad-headed lizard,Phrynocephalus vlangalii. The study area is along a 20 km stretch of the railway on northern Qinghai-Tibet Plateau,and this section of the railway was constructed between 1958–1979. Both assignment tests and analysis of molecular variance(AMOVA) were used for data analysis. We found significant genetic differentiation between the populations from the study area and those from a further southeastern area,which are separated by a 20 km gap. This suggests the existence of population substructure at a fine-scale. However,we did not detect any difference between samples from the western and eastern sides of the railway within the study area,and concluded that the railway may not impose a significant barrier effect on these lizard populations at the present time. Available suitable habitat alongside the railway and bridge underpasses may have facilitated the gene exchange between the sides. The relatively short time since the completion of the railway may not allow the differentiation to accumulate to a detectable level. Since the Qinghai-Tibet Plateau maintains a unique and fragile ecosystem,long-term monitoring of such man-made landscape features is imperative for protecting this ecosystem.

Clustering-based maintainability demonstration for complex systems with a mixed maintenance time distribution

During maintainability demonstration,the maintenance time for complex systems consisting of mixed technologies generally conforms to a mixture distribution.However existing maintainability standards and guidance do not explain explicitly how to deal with this situation.This paper develops a comprehensive maintainability demonstration method for complex systems with a mixed maintenance time distribution.First of all,a K-means algorithm and an expectation-maximization(EM)algorithm are used to partition the maintenance time data for all possible clusters.The Bayesian information criterion(BIC)is then used to choose the optimal model.After this,the clustering results for equipment are obtained according to their degree of membership.The degree of similarity for the maintainability of different kinds of equipment is then determined using the projection method.By using a Bootstrap method,the prior distribution is obtained from the maintenance time data for the most similar equipment.Then,a test method based on Bayesian theory is outlined for the maintainability demonstration.Finally,the viability of the proposed approach is illustrated by means of an example.

Towards Fast and Efficient Algorithm for Learning Bayesian Network

Learning Bayesian network structure is one of the most exciting challenges in machine learning. Discovering a correct skeleton of a directed acyclic graph（DAG） is the foundation for dependency analysis algorithms for this problem. Considering the unreliability of high order condition independence（CI） tests, and to improve the efficiency of a dependency analysis algorithm, the key steps are to use few numbers of CI tests and reduce the sizes of conditioning sets as much as possible. Based on these reasons and inspired by the algorithm PC, we present an algorithm, named fast and efficient PC（FEPC）, for learning the adjacent neighbourhood of every variable. FEPC implements the CI tests by three kinds of orders, which reduces the high order CI tests significantly. Compared with current algorithm proposals, the experiment results show that FEPC has better accuracy with fewer numbers of condition independence tests and smaller size of conditioning sets. The highest reduction percentage of CI test is 83.3% by EFPC compared with PC algorithm.