On MLEs of the parameters of a modified Weibull distribution for progressively type-2 censored samples

Lifetimes of modern mechanic or electronic units usually exhibit bathtub-shaped failure rates. An appropriate probability distribution to model such data is the modified Weibull distribution proposed by Lai et al. [15]. This distribution has both the two-parameter Weibull and type-1 extreme value distribution as special cases. It is able to model lifetime data with monotonic and bathtub-shaped failure rates, and thus attracts some interest among researchers because of this property. In this paper, the procedure of obtaining the maximum likelihood estimates (MLEs) of the parameters for progressively type-2 censored and complete samples are studied. Existence and uniqueness of the MLEs are proved.     

Search boundaries of the truncated discrete sequential test

The theme of this paper is improved planning of binomial sequential probability ratio tests in the context of comparison of two objects as to their time between failures or to failure, assumed to be exponentially distributed. The authors’ earlier works established that the probabilities of I- and II- type errors (α and β) are discrete in character and do not lend themselves to analytical expression. Accordingly, the choice of the optimal parameters for the decision boundaries necessitates a search for extrema in discrete sets. The present work outlines a procedure that involves application of the continued-fractions theory, and permits finding the set of boundary positions in which the test characteristics undergo changes. It was established, that in the domains described in the earlier papers, the relationships of α and β versus these positions are close to planar and – within narrow limits – stepwise. The step sizes are highly variable, so that the standard minimum search procedures are either cumbersome or actually useless. On the basis of these relationships~– and others – a search algorithm is proposed for the optimal test boundaries. An example is presented – planning and implementation of this test in the integrated-circuit industry.     

Detection of changes in a random financial sequence with a stable distribution

Quick detection of unanticipated changes in a financial sequence is a critical problem for practitioners in the finance industry. Based on refined logarithmic moment estimators for the four parameters of a stable distribution, this article presents a stable-distribution-based multi-CUSUM chart that consists of several CUSUM charts and detects changes in the four parameters in an independent and identically distributed random sequence with the stable distribution. Numerical results of the average run lengths show that the multi-CUSUM chart is superior (robust and quick) on the whole to a single CUSUM chart in detecting the shift change of the four parameters. A real example that monitors changes in IBM's stock returns is used to demonstrate the performance of the proposed method.     

New Tests of Spatial Segregation Based on Nearest Neighbour Contingency Tables

Abstract.  The spatial clustering of points from two or more classes (or species) has important implications in many fields and may cause segregation or association, which are two major types of spatial patterns between the classes. These patterns can be studied using a nearest neighbour contingency table (NNCT) which is constructed using the frequencies of nearest neighbour types. Three new multivariate clustering tests are proposed based on NNCTs using the appropriate sampling distribution of the cell counts in a NNCT. The null patterns considered are random labelling (RL) and complete spatial randomness (CSR) of points from two or more classes. The finite sample performance of these tests are compared with other tests in terms of empirical size and power. It is demonstrated that the newly proposed NNCT tests perform relatively well compared with their competitors and the tests are illustrated using two example data sets.     

A Model Validation Procedure when Covariate Data are Missing at Random

Abstract. In the presence of missing covariates, standard model validation procedures may result in misleading conclusions. By building generalized score statistics on augmented inverse probability weighted complete‐case estimating equations, we develop a new model validation procedure to assess the adequacy of a prescribed analysis model when covariate data are missing at random. The asymptotic distribution and local alternative efficiency for the test are investigated. Under certain conditions, our approach provides not only valid but also asymptotically optimal results. A simulation study for both linear and logistic regression illustrates the applicability and finite sample performance of the methodology. Our method is also employed to analyse a coronary artery disease diagnostic dataset.     

An RKHS framework for functional data analysis

Linear combinations of random variables play a crucial role in multivariate analysis. Two extension of this concept are considered for functional data and shown to coincide using the Loève–Parzen reproducing kernel Hilbert space representation of a stochastic process. This theory is then used to provide an extension of the multivariate concept of canonical correlation. A solution to the regression problem of best linear unbiased prediction is obtained from this abstract canonical correlation formulation. The classical identities of Lawley and Rao that lead to canonical factor analysis are also generalized to the functional data setting. Finally, the relationship between Fisher's linear discriminant analysis and canonical correlation analysis for random vectors is extended to include situations with function-valued random elements. This allows for classification using the canonical Y scores and related distance measures.     

用于分类的随机森林和Bagging分类树比较

借助试验数据,从两种理论分析角度解释随机森林算法优于Bagging分类树算法的原因。将两种算法表述在两种不同的框架下,消除了这两种算法分析中的一些模糊之处。尤其在第二种分析框架下,更能清楚的看出,之所以随机森林算法优于Bagging分类树算法,是因为随机森林算法对应更小的偏差。     

服务补救中顾客情绪对顾客满意之影响的实证研究

在顾客服务补救满意的研究中,本文引入情绪变量,提出了包含服务失败、服务补救整个过程的研究模型,采用情景模拟的方法进行了实证研究。研究表明:顾客在服务补救后的满意,受顾客服务补救后情绪的影响,而不是受顾客服务失败时的初始情绪的影响。其中,补救后积极情绪对补救满意有显著的正向影响;补救后外在归因消极情绪对补救满意有显著的负向影响。     

Asymptotic expansions and the reliability of tests in accelerated failure time models

This paper gives matrix formilae for the O(n^-1 ) cerrecti0n applicable to asymptotically efficient conditional moment tests. These formulae only require expectations of functions involving, at most, second order derivatives of the log-likelihood; unlike those previously providcd by Ferrari and Corddro(1994). The correction is used to assess the reliability of first order asymptotic theory for arbitrary residual-based diagnostics in a class of accelerated failure time models: this correction is always parameter free, depending only on the number of included covariates in the regression design. For all but one of the tests considered, first order theory is found to be extremely unreliable, even in quite large samples, although this may not be widely appreciated by applied workers.     

An empirical bias–variance analysis of DECORATE ensemble method at different training sample sizes

DECORATE (Diverse Ensemble Creation by Oppositional Relabeling of Artificial Training Examples) is a classifier combination technique to construct a set of diverse base classifiers using additional artificially generated training instances. The predictions from the base classifiers are then integrated into one by the mean combination rule. In order to gain more insight about its effectiveness and advantages, this paper utilizes a large experiment to study the bias–variance analysis of DECORATE as well as some other widely used ensemble methods (such as bagging, AdaBoost, random forest) at different training sample sizes. The experimental results yield the following conclusions. For small training sets, DECORATE has a dominant advantage over its rivals and its success is attributed to the larger bias reduction achieved by it than the other algorithms. With increase in training data, AdaBoost benefits most and the bias reduced by it gradually turns to be significant while its variance reduction is also medium. Thus, AdaBoost performs best with large training samples. Moreover, random forest behaves always second best regardless of small or large training sets and it is seen to mainly decrease variance while maintaining low bias. Bagging seems to be an intermediate one since it reduces variance primarily.