Communications in Statistics: Theory and Methods

Yang, Jianfeng; Liu, Min-Qian; Zhang, Runchu

doi: 10.1080/03610920802618418pmid: N/A

Fractional factorial split-plot (FFSP) designs have received much attention in recent years. In this article, the matrix representation for FFSP designs with multi-level factors is first developed, which is an extension of the one proposed by Bingham and Sitter (1999b) for the two-level case. Based on this representation, periodicity results of maximum resolution and minimum aberration for such designs are derived. Differences between FFSP designs with multi-level factors and those with two-level factors are highlighted.

Yang, Jingyun; Chinchilli, Vernon M.

doi: 10.1080/03610920802618426pmid: N/A

Cohen's kappa statistic is the conventional method that is used widely in measuring agreement between two responses when they are categorical. In this article, we develop a fixed-effects modeling of Cohen's kappa for bivariate multinomial data which reduces to Cohen's kappa under certain conditions and hence can be considered as a generalization of the conventional Cohen's kappa. Also, this method can easily be adapted as a generalization of Cohen's weighted kappa. Properties of the proposed method are provided. Large sample performance is investigated through bootstrap simulation studies followed by two illustrative examples.

Wang, Kaiping; Lin, Lu; Qi, Ruihua

doi: 10.1080/03610920802632526pmid: N/A

In this article, we extend a semiparametric regression estimator with multiplicative adjustment to time series context. The asymptotic theory and results from a simulation study are discussed. Theoretical results and numerical comparison show that, in the time series case, the semiparametric estimator is better than the traditional local polynomial estimator in a wide neighbourhood around the true regression function.

Liu, Rong; Ramakrishnan, Viswanathan

doi: 10.1080/03610920802641709pmid: N/A

Many longitudinal clinical studies suffer from nonignorable dropouts. Ramakrishnan and Wang (2005) proposed a mixed effects analysis treating missing data as missing due to truncation (MDT), estimated the parameters under a multivariate truncated normal model, and suggested an adjustment to the degrees of freedom to account for the implicit estimation of the missing data. We propose a multiple imputation (MI) in conjunction with the MDT method as an alternative to accurately accommodate the uncertainty introduced by imputation. The data used in Ramakrishnan and Wang (2005) is considered for illustration. A comparison of various methods using a simulation study is presented.

Li, Jiexiang

doi: 10.1080/03610920802641717pmid: N/A

Consider observations (representing lifelengths) taken on a random field indexed by lattice points. Estimating the distribution function F(x) = P(X i  ≤ x) is an important problem in survival analysis. We propose to estimate F(x) by kernel estimators, which take into account the smoothness of the distribution function. Under some general mixing conditions, our estimators are shown to be asymptotically unbiased and consistent. In addition, the proposed estimator is shown to be strongly consistent and sharp rates of convergence are obtained.

Iqbal, Ijaz; Tahir, M. H.

doi: 10.1080/03610920802642566pmid: N/A

Magda (1980) and Hedayat (1981) first considered the construction of circular strongly balanced repeated measurements designs. Sen and Mukerjee (1987) and Roy (1988) considered the optimality and existence of circular strongly balanced repeated measurements designs based on the method of differences and Hamiltonian decomposition of lexicographic product of two graphs. In this article, we consider the construction of circular strongly balanced repeated measurements designs using the newly proposed method called cyclic shifts, and propose some new designs for p < v.

Seck, Cheikh Tidiane; Lo, Gane Samb

doi: 10.1080/03610920802645387pmid: N/A

We consider in this note the weak convergence, in the frame of the empirical processes theory, of the nonweighted poverty measures viewed as stochastic processes defined on some space of bounded functions and indexed by real numbers or monotone functions. The results include the asymptotic behavior of the Foster–Greer–Thorbecke process of poverty indices. We use them to follow up the poverty evolution in poor countries between two periods with appropriate curves.

Duan, Xiaogang; Liu, Liping; Zhao, Ping

doi: 10.1080/03610920802645395pmid: N/A

The main purpose of this article is to estimate the underlying covariate distribution with a biased capture–recapture sample. Two procedures are proposed, and the derived estimates are found to be consistent and asymptotically normal. The proposed methods are compared and are shown to perform well in most circumstances via simulation study, and are applied to a real example.

Liang, Tachen; Huang, Wen-Tao

doi: 10.1080/03610920802645403pmid: N/A

This article deals with the empirical Bayes estimation of the parameter θ in a uniform distribution U(0, θ) based on randomly right censored data. By mimicking the form of the Bayes estimator, an empirical Bayes estimator is constructed. The asymptotic optimality of is investigated. It is shown that under certain conditions, is asymptotically optimal with a rate of convergence n −λr/2(r+1), where n is the number of past data available when the present estimation problem is considered, and 0 < λ ≤2, and r is a positive integer related to some conditions.

Chai, Feng-Shun; Chatterjee, Kashinath; Gupta, Sudhir

doi: 10.1080/03610920802650320pmid: N/A

Several extensions of the popular E(s 2) criterion of Booth and Cox (1962) to multilevel supersaturated designs have been advanced in literature. These extensions are not unique due to different ways they measure overall nonorthogonality between all pairs of the columns of the model matrix. We exploit the connection of the E(s 2) criterion with A- and D-optimality that naturally lends itself to a generalized criterion for the multilevel situation in a unified way. The extensions provided in literature follow as special cases of the generalized criterion. A lower bound to the generalized criterion is derived for a wide class of designs, and a method of construction for the symmetrical case is discussed.