摘要:

Five tests for the hypothesis of no treatment effects in complete blocks designs are compared with respect to asymptotic relative efficiency: the ordinary two-way analysis of variance, the analysis of variance of ranks (Friedman, 1937), the method of ranking after alignment (Hodges and Lehmann, 1962), and a method of weighted rankings proposed by Quade (1972, 1979). This comparison required devising a procedure for estimating the asymptotic relative efficiencies of test statistics with asymp-totic chi-squared distributions.

ISSN:0361-0918    

DOI:10.1080/03610918308812339

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

In this study, the robustness of power and significance level of several statistical testing methods was evaluated under the assumption that the test populations were from Poisson, negative binomial, or geometric distributions. The F-ratio test, with or without appropriate transformations, was shown to be both safe and robust for all distributions examined.

ISSN:0361-0918    

DOI:10.1080/03610918308812340

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

We describe the design and analysis for a simulation experiment to compare the mean-squared errors (MSE's) of two quantile estimators defined for random walk designs. The dependence of the easily computed MSE of the first estimator on the levels of five factors is examined via multiple regression. This information is used to plan a simulation to compute the MSE of the second estimator using a fraction of a 3352factorial allowing uncorrelated estimates for all main effects and the two-factor interactions of a specified factor. Efficient estimation of the MSE of the second estimator is attempted through antithetic and control variate techniques of variance reduction, with modest success.

ISSN:0361-0918    

DOI:10.1080/03610918308812341

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

The bounds of Birnbaum (1942) and Sampford (1953) for the upper tail area of the normal distribution are extended to the upper tail of the t-distribution. Numerical and theoretical comparisons are made with the bounds of Peizer and Pratt (1968), Wallace (1959) and Soms (1977).

ISSN:0361-0918    

DOI:10.1080/03610918308812342

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

The paper introduces a x2-approximation to multivariate kurtosis b2,punder normality. It requires calculating the third moment of b2,pwhich is obtained. We compare the approximation with simulated percentage points and the normal approximation, and find it to be adequate for p=l and 2. For p=3, the simple average of this estimate and the normal approximation is found to be generally superior to either approximation on its own. For p=4, the normal approximation is best for non-extreme values of ∝

ISSN:0361-0918    

DOI:10.1080/03610918308812343

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

A Sampling experiment performed using data collected for a large clinical trial shows that the discriminant function estimates of the logistic regression coefficients for discrete variables may be severely biased. The simulations show that the mixed variable location model coefficient estimates have bias which is of the same magnitude as the bias in the coefficient estimates obtained using conditional maximum likelihood estimates but require about one-tenth of the computer time.

ISSN:0361-0918    

DOI:10.1080/03610918308812344

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

A mixed-integer programing formulation for clustering is proposed, one that encompasses a wider range of objectives--and side conditions--than standard clustering approaches. The flexibility of the formulation is demonstrated in diagrams of sample problems and solutions. Preliminary computational tests in a practical setting confirm the usefulness of the formulation.

ISSN:0361-0918    

DOI:10.1080/03610918308812345

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

This paper presents the small sample optimum choice of k ≤n + r1− r2+ 1) order statistics for the best linear unbiased estimates (BLUES) of the parameters μ and σ or σ alone ( μ known) when the sample is Type II censored in the middle retaining only r1lower and n - r2+ 1 upper order statistics. For n = 3(1)10, k = 2(1)4, r1= O(1) (n−2) and r2= (r1+2) (l)n, the optimum ranks, the coefficients of the BLUEs have been presented in Table I

ISSN:0361-0918    

DOI:10.1080/03610918308812346

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

摘要:

Some clarification of statistics based on McKay's x2approximation for the distribution of the sample coefficient of variation is presented. The conclusions of Warren (1982) are shown to result from the confusion of two definitions for the sample coefficient of variation.

ISSN:0361-0918    

DOI:10.1080/03610918308812347

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610918308812348

出版商:Marcel Dekker, Inc.    

年代:1983

数据来源: Taylor