摘要:

The purpose of screening is the reduction of a large set of explanatory variables or factors to the set of important variables, assuming that there are only a few really important explanatory variables. The technique for screening that we propose is a modification of Jacoby and Harrison's (1962) sequential bifurcation, and resembles binary search. It assumes that the direction of the effect of potential variables is already known.This paper starts with deterministic linear response surfaces, for ease of description and easy comparison with alternative techniques. A large scale application illustrates the method. Next the technique is extended with normally distributed random errors with a known common standard deviation. Simulation results include the number of (correctly) found important variables, the number of (incorrectly) found unimportant variables, and the number of observations needed. Further extensions and limitations are briefly indicated.

ISSN:0361-0918    

DOI:10.1080/03610919508813236

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

A life test plan(LTP)is developed for testing the exponential mean lifetime in the Type I censoring case. For a given censoring time, the required sample size and the critical region are determined either numerically or by Monte Carlo simulation such that the producer and the consumer risks are satisfied. In addition the developed LTP is compared to the hybrid LTP in terms of sample size, expected completion time, and power. Computational results indicate that the former generally requires a smaller sample size and less completion time on the averge than the latter. As for the power, neither plan completely dominates the other at all values of the true mean lifetime

ISSN:0361-0918    

DOI:10.1080/03610919508813237

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

Suppose that the quadratic logistic model describes the relationship between a proportion p and some independent variable x. When constructing interval estimates for the value of x at which p attains an optimum value, the following asymptotic methods are available: 1) Fieller; 2) delta; 3) bootstrap variance; 4) bootstrap percentile; 5) bootstrap studentized percentile; and 6) jackknife. It is not known, however, how these asymptotic methods compare for finite sample sizes. For balanced D-optimum designs with equally spaced design points, simulation results were used to compare the small and moderate sample-size performance of these asymptotic methods. Overall, in terms of coverage probabilities and interval length characteristics, the delta method and bootstrap studentized percentile method gave the best results for small and moderate sample sizes, respectively.

ISSN:0361-0918    

DOI:10.1080/03610919508813238

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

The Fisher distribution is a commonly employed model for unimodal data on the surface of a sphere. The sampling distribution of the resultant is needed for making inferences regarding its concentration parameter. Available approximations to the sampling distribution are appropriate only over restricted ranges of the parameter. We propose a two moment beta approximation that is accurate for a wide range of sample sizes and parameter values of practical interest. It is convenient for testing hypotheses and obtaining confidence intervals; nomograms useful for such purposes are included.

ISSN:0361-0918    

DOI:10.1080/03610919508813239

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

Data from yield trials conducted in different environments are frequently analysed by ordinary analysis of variance. One of the preconditions for such analyses is that error variances be homogeneous across treatments and replicates. If a mixed model with fixed genotypes and random environments is assumed, genotype-environmental interaction is a random effect. For the ordinary ANOVA to be valid in this case, it is then also required that interaction variances be homogeneous for different genotypes. This paper suggests Box's correction of the ANOVA degrees of freedom for the case of heterogeneous interaction and error variances Different estimators of the correction factors are compared via Monte carlo Simulation.

ISSN:0361-0918    

DOI:10.1080/03610919508813240

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

The equations for the solution of minimum variance stratification under Neyman allocation are not easily tractable and hence approximate solutions have been suggested by various workers. These approximate solutions are in general not satisfactory as they do not achieve global minimum variance. An iterative procedure which results in global minimum variance solution to the stratification problem under Neyman allocation, is suggested herein with illustrations using six theoretical population models.

ISSN:0361-0918    

DOI:10.1080/03610919508813241

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610919508813225

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor