摘要:

A technique for estimating the quantiles or percentiles of a distribution is developed. The parametric form of the distribution is assumed unknown. The estimation procedure is based on a kernel estimator of a probability density function and on aquantile estimator suggested by Harrell and Davis (1982). Simulation studies show that estimation of quantiles in moderately heavyto heavy tails of a distribution is substantially improved by use of the technique.

ISSN:0361-0918    

DOI:10.1080/03610918708812592

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

Cox's discrete logistic model was extended to the study of the life table by Thompson (1977) to handle grouped survival data. Inferences about the effect of grouping are studies byMonte Carlo methods. The results show that the effect of grouping is not substantial. This approach is applied to the grouped data on liver cancer. The computer program developed for grouped censored data with continuous and indicator covariates is of practical importance and is available fromThe Ohio State University

ISSN:0361-0918    

DOI:10.1080/03610918708812593

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

Quade (1972, 1979) proposed a family of nonparametric tests based on weighted within-block rankings, for testing the hypothesis of no treatment effects in a complete randomized blocks layout. In this paper we give a table of the exact null distribution of these tests when the number of treatments is 3, the number of blocks is less than or equal to 14 and the block scores are linear. Moreover, a Monte Carlo study was performed to compare the powers of these tests with parametric and nonparametric competitors

ISSN:0361-0918    

DOI:10.1080/03610918708812594

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

Results of the Monte Carlo study of the performance of a maximum likelihood estimation in a Weibull parametric regression model with two explanatory variables are presented. One simulation run contained 1000 samples censored on the average by the amount of 0-30%. Each simulatedsample was generated in a form of two-factor two-level balanced experiment. The confidence intervals were computed using the large-sample normal approximation via the matrix of observed information. For small sample sizes the estimates of the scale parameter b of the loglifetime were significantly negatively biased, which resulted in a poor quality of confidence intervals for b and the low-level quantiles. All estimators improved their quality when the nominal value of b decreased. A moderate amount of censoring improved the quality of point and confidence estimation. The reparametrization b 7 produced rather accurate confidence intervals. Exact confidence intervals for b in case of non-censoring were obtained using the pivotal quantity b/b.

ISSN:0361-0918    

DOI:10.1080/03610918708812595

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

The generalised least squares, maximum likelihood, Bain-Antle 1 and 2, and two mixed methods of estimating the parameters of the two-parameter Weibull distribution are compared. The comparison is made using (a) the observed relative efficiency of parameter estimates and (b) themean squared relative error in estimated quantiles, to summarize the results of 1000 simulated samples of sizes 10 and 25. The results are that: generalised least squares is the best method of estimating the shape parameter ß the best method of estimating the scale parameter a depends onthe size of ß for quantile estimation maximum likelihood is best Bain-Antle 2 is uniformly the worst of the methods.

ISSN:0361-0918    

DOI:10.1080/03610918708812596

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

A unifying perspective on variance reduction is presented that emphasizes broadly defined variance reduction strategies rather than specific variance reduction techniques (VRTs). The perspective is based on a new taxonomy of VRTs, which is reviewed in detail. The variance reduction problem is formulated as a constrained optimization problem, and results that guarantee the effectiveness of variance reduction strategies are summarized.

ISSN:0361-0918    

DOI:10.1080/03610918708812597

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

This paper presents a graphical procedure for simultaneously distinguishing between two commonly encountered data anomaliesWhen applied in the context of one anomaly, a family of parallel lines will be estimated, and when applied in the contextofthe second anomaly, a family of lines, whose members all pass through the same point, will be estimated. It is shown that the procedure can be applied effectively using samples containing as few as two hundred bivariate observations.

ISSN:0361-0918    

DOI:10.1080/03610918708812598

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

It is generally considered that analysis of variance by maximum likelihood or its variants is computationally impractical, despite existing techniques for reducing computational effect per iteration and for reducing the number of iterations to convergence. This paper shows thata major reduction in the overall computational effort can be achieved through the use of sparse-matrix algorithms that take advantage of the factorial designs that characterize most applications of large analysis-of-variance problems. In this paper, an algebraic structure for factorial designsis developed. Through this structure, it is shown that the required computations can be arranged so that sparse-matrix methods result in greatly reduced storage and time requirements.

ISSN:0361-0918    

DOI:10.1080/03610918708812599

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

Five tests of homogeneity for a 2x(k+l) contingency table are compared using Monte Carlo techniques. For these studiesit is assumed that k becomes large in such a way that thecontingency table is sparse for 2xk of the cells, but the sample size in two of the cells remains large. The test statistics studied are: the chi-square approximation to the Pearson test statistic, the chi-square approximation to the likelihood ratio statistic, the normal approximation to Zelterman's (1984)the normal approximation to Pearson's chi-square, and the normal approximation to the likelihood ratio statistic. For the range of parameters studied the chi-square approximation to Pearson's statistic performs consistently well with regard to its size and power.

ISSN:0361-0918    

DOI:10.1080/03610918708812600

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor

摘要:

The sample linear discriminant function (LDF) is known to perform poorly when the number of features p is large relative to the size of the training samples, A simple and rarely applied alternative to the sample LDF is the sample Euclidean distance classifier (EDC). Raudys and Pikelis (1980) have compared the sample LDF with three other discriminant functions, including thesample EDC, when classifying individuals from two spherical normal populations. They have concluded that the sample EDC outperforms the sample LDF when p is large relative to the training sample size. This paper derives conditions for which the two classifiers are equivalent when all parameters are known and employs a Monte Carlo simulation to compare the sample EDC with the sample LDF no only for the spherical normal case but also for several nonspherical parameter configurations. Fo many practical situations, the sample EDC performs as well as or superior to the sample LDF, even for nonspherical covariance configurations.

ISSN:0361-0918    

DOI:10.1080/03610918708812601

出版商:Marcel Dekker, Inc.    

年代:1987

数据来源: Taylor