摘要:

This paper considers lifetime or other regression models in which the response variable may be grouped or censored. We study a particular family that includes logistic and extreme value regression models and find that grouping or censoring often has only a moderately harmful effect on the efficiency of estimation of regression coefficients. On the other hand, the effects are drastic with regard to attempts to discriminate among alternative regression error distributions. Practical consequences of the results are discussed

ISSN:0361-0918    

DOI:10.1080/03610918808812688

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

Several biased estimators have been proposed as alternatives to the least squares estimator when multicollinearity is present in the multiple linear regression model. The ridge estimator and the principal components estimator are two techniques that have been proposed for such problems. In this paper the class of fractional principal component estimators is developed for the multiple linear regression model. This class contains many of the biased estimators commonly used to combat multicollinearity. In the fractional principal components framework, two new estimation techniques are introduced. The theoretical performances of the new estimators are evaluated and their small sample properties are compared via simulation with the ridge, generalized ridge and principal components estimators

ISSN:0361-0918    

DOI:10.1080/03610918808812689

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

The purpose of this paper is two-fold. One is to compare the almost unbiased generalized ridge regression (AUGRR) estimator proposed by Singh, Chaubey and Dwivedi (1986) with the generalized ridge regression (GRR) estimator and with the ordinary least squares (OLS) estimator in terms of the mean squared error criterion. Second is to examine small sample properties of the operational almost unbiased ordinary ridge regression (AUORR) estimator by Monte Carlo experiments.

ISSN:0361-0918    

DOI:10.1080/03610918808812690

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

The systematic error (bias) of the isotonic regression analysis of temporal spacings between failure events is investigated by means of numerical simulation. Spacings that are sampled from an exponential distribution with a constant failure rate (CFR) arc subjected to an isotonic regression search for a declining failure rate (DFR). The results indicate a considerable declining trend (bias) that is imposed upon these CFR-data by isotonic regression analysis. The corresponding results for an increasing trend can be readily obtained through transformation. For practical applications, the results of 100,000 simulations have been approximated by simple analytical expressions. For the evaluation of a trend in a specific set of isotonized spacings (or rates) the results of the latter analysis can be compared with the isotonic bias of a set of CFR data for the same number of events. Alternatively, the specific set of isotonized spacings can be suitably related to the corresponding isotonized CFR data to reduce the bias by largely eliminating the CFR contribution.

ISSN:0361-0918    

DOI:10.1080/03610918808812691

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

The set of all distinct blocks of a BIBD(v,b,r,k,λ) is referred to as the support of the design. In this paper, the family of BIB designs with v=9 and k=3 is studied from the view of possible support sizes, b*'s. A table is constructed of designs with support sizes belonging to {12,18,20,21,…,84}, for minimum possible b in each case and for any larger admissible b. In constructing this table the methods of trade-off and composition of designs are utilized

ISSN:0361-0918    

DOI:10.1080/03610918808812692

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

Optimal 4x4 square and nonsquare plot designs are given under two different correlation structures. The A and D criteria for optimality are used. Nonsquare plots are shown to be superior to square plots of equal size under certain conditions when correlations are present. The optimal designs are not generally identical under both correlation structures

ISSN:0361-0918    

DOI:10.1080/03610918808812693

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

If events are scattered in Rnin accordance with a homogeneous Poisson process and if X is the location of the event with minimal [d]lPnorm, then in the case p = n the nthabsolute powers of the coordinates of X form a sample of size n from a gamma distribution with shape parameter 1/n. In an age of parallel computing, this fact may lead to some attractive simulation methods. One possibility is to generate R = [d]X[d] and U = Y/[d]X[d] independently, perhaps by setting U = Y/[d]Y[d] where Y has any p.d.f. which is a function only of ¦Y¦. We consider for example Y having the uniform distribution in an lPball.

ISSN:0361-0918    

DOI:10.1080/03610918808812694

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

Kurtosis, usually as measured by the standardised fourth central moment, has been examined on a number of occasions by observing the effect of contaminating the distribution, that is, mixing in another distribution. However, superficial treatment can lead, and indeed has led, to misunderstandings. This paper considers, firstly for a symmetric distribution contaminated at two points symmetrically placed around its centre and then for a mixture of two continuous symmetric distributions, the behaviour of three measures of kurtosis. This is done in general and not just as the mixing proportion tends to zero as in the influence function approach. It is seen that when both scale and kurtosis change, the latter is not necessarily intuitive. It is also illustrated that parameter interpretation in terms of distributional properties such as shape can be misleading without the use of the appropriate distributional partial ordering

ISSN:0361-0918    

DOI:10.1080/03610918808812695

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

A Monte Carlo study has been performed to compare the powers of various asymptotically optimal tests for testing exponentiality against positive aging. It is observed that the test based on sample coefficient of variation performs better than its competitors in the various cases

ISSN:0361-0918    

DOI:10.1080/03610918808812696

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor

摘要:

Epidemiologic data are often summarized in the form of several 2x2 tables where it is of interest to estimate the odds ratio for each 2x2 table. More importantly these estimators are often combined into a single estimator and a test statistic for the hypothesis of no association computed. The performance of three such combined odds ratio estimators and four tests of association was studied using Monte Carlo methods. The results of these studies are described in this paper. For the Monte Carlo studies, a constant odds ratio was used with equal numbers of cases and controls. A wide range of odds ratios, probabilities of exposure, number of cases, and number of strata were used. For each of the 1000 Kx2x2 tables used in one simulation, the Mantel-Haenszel (1959), maximum conditional likelihood (Gart, 1971), and weighted least squares (Woolf, 1955) estimators of the odds ratio were computed along with the likelihood-ratio (Bishop et al., 1975), Mantel-Haenszel, Pearson, and weighted least squares tests of association. The results of these studies indicate that the mean squared error of the weighted least squares estimator is usually less than the mean squared Pearson statistic (9), since it may take negative values. It was included in this simulation since it is a statistic produced by some software packages. The LR and MH tests are recommended since they maintain their size near 5% over the range of parameters used in the simulation.

ISSN:0361-0918    

DOI:10.1080/03610918808812697

出版商:Marcel Dekker, Inc.    

年代:1988

数据来源: Taylor