ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480528

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

The assessment of goodness-of-fit for logistic regression models using categorical predictors is made complicated by the fact that there are different ways of defining the saturated model. Three distinct approaches have been suggested: the casewise approach, the contingency table approach, and the collapsing approach. Assessments of goodness-of-fit using χ2statistics are very different depending on which of these approaches is used, and different statistical packages do different things (often depending on how the data are entered into the package). In this article the different approaches are outlined, and the functionality of different statistical packages is described. Cautions for the unwary data analyst are offered, and suggestions for improvements to statistical software that would make logistic regression analysis easier for the data analyst are made. Some advice for data analysts using current versions of such software is also provided.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480529

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

Many textbooks on regression analysis include the methodology of principal components regression (PCR) as a way of treating multicollinearity problems. Although we have not encountered any strong justification of the methodology, we have encountered, through carrying out the methodology in well-known data sets with severe multicollinearity, serious actual and potential pitfalls in the methodology. We address these pitfalls as cautionary notes, numerical examples that use well-known data sets. We also illustrate by theory and example that it ispossiblefor the PCR to fail miserably in the sense that when the response variable is regressed on all of thepprincipal components (PCs), the first (p− 1) PCs contribute nothing toward the reduction of the residual sum of squares, yet the last PC alone (the one that is always discarded according to PCR methodology) contributes everything. We then give conditions under which the PCR totally fails in the above sense.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480530

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

Polynomial regression is a common technique for estimating conditional mean functions of unknown form. Increasing the order of the polynomial increases the flexibility of the estimated function. This note shows with an example how these estimated polynomials, even (perhaps especially) higher-order ones, display undesirable nonlocal effects. That is, values of the response corresponding to a particular value or range of the predictor may have a large and undesirable influence on the predicted response at a very different value of the predictor.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480531

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

Data from a cement example published by Hald has been widely used in the statistics literature to illustrate collinearity and variable selection methods. Study of the 1932 article from which Hald obtained the data reveals several interesting facts: (1) the data resulted from a mixture experiment; (2) the collinearity among the independent variables resulted from transforming oxide compositions to compound compositions; (3) Hald does not list or discuss one component of the cement mixture that is essentially constant for all data points except one; and (4) this point is thus potentially influential. The cement data published in the 1932 article are listed and discussed. Mixture experiment analyses for cement compositions expressed in both oxide and compound components are presented and compared. The mixture experiment analyses are also compared to typical nonmixture experiment analyses of the data. Based on this work, we recommend that textbook, journal, and instructional uses of the Hald cement data set take into account the four facts listed, and analyze the data using mixture experiment methods. We also make several general recommendations that data analysts should consider when faced with other data sets involving mixtures or independent variable transformations.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480532

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

Over the past four years, the U.S. Office of Management and Budget (OMB) has conducted a comprehensive review of the statistical standard that governs federal agency collection and publication of data on race and ethnicity. The principal objective of this review was to enhance the accuracy of demographic information produced by the federal government. The multiyear process had two essential elements—public comment on the standard that had been in place since 1977, as well as on suggestions that had been made for changing the standard, and research and testing to assess the possible effects of suggested changes on the quality and utility of data on race and ethnicity. This article presents highlights of the review process and the decisions on modifying the standard that OMB announced in October 1997.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480533

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480534

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

An annotated list is presented containing presumed first occurrences in print of terms commonly used in probability and statistics. The list supplements and provides some corrections to a longer list published in volume 49 ofThe American Statistician.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480535

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

It is shown that for any finite population of sizeNand any sample sizen≥ 3, we can have a survey plan of sizenthat assigns a probability of zero to an undesirable sample of sizenand yet the plan is equivalent to a simple random sample plan of sizenwithout replacement.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480536

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

This article describes an idea for motivating students in an introductory probability and statistics course. The motivating theme isrisk, and the process begins with a first-day-of-class questionnaire that samples attitudes of students toward risk and involves them in analysis of events and decisions from their daily lives. Questionnaire responses serve as a context for the instructor to develop the technical concepts of probability and statistics. Moreover, the questionnaire provides a way to increase substantially student motivation and involvement in the course.

ISSN:0003-1305    

DOI:10.1080/00031305.1998.10480537

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor