# Journal of the American Statistical Association

Journal covers statistical science and its applications, theory, and methods in economic, social, physical, engineering and health sciences.

PublicationArticlesSingle

## Articles from Vol. 84, No. 408, December

A Distribution-Free Multivariate Sign Test Based on Interdirections
1. INTRODUCTION Let [X.sub.j] = ([X.sub.1j], ..., [X.sub.pj])' (j = 1, ..., n) denote a sample from a p-dimensional, absolutely continuous population with a p-vector location parameter [theta]. The population is assumed to be angularly symmetric...
Adjusting the 1980 Census of Population and Housing
Nearly the whole of the states have now returned their census. I send you the result, which as far as founded on actual returns is written in black ink, and the numbers not actually returned, yet pretty well known, are written in red ink. Making a...
A Markov Mixture Model for Magazine Exposure
1. INTRODUCTION Suppose an advertiser is about to launch an advertising campaign by placing one advertisement in k successive issues of a magazine. One of his objectives is to estimate the proportion of the population that sees none, one, or as...
1. INTRODUCTION This article develops inference procedures for the highest-order interaction in unreplicated multiway classifications. The procedures require that exactly one observation be made under each treatment combination. The procedures could...
Bayesian Estimation and Prediction for Pareto Data
1. INTRODUCTION Consider data from a classical Pareto distribution with shape, or inequality, parameter [alpha] and precision parameter [tau]. The corresponding density is f(x; [alpha], [tau]) = [tau][alpha]([tau]x[).sup.-([alpha] + 1), [tau]x...
Bayesian Marginal Inference
1. DESCRIPTION OF PROBLEM Bayesian methodology provides a general paradigm for inference regarding any function of the unknown parameters in nonlinear models, even when the sample size is finite, and whether or not prior information is available....
Binomial Regression with Monotone Splines: A Psychometric Application
1. THE BINOMIAL REGRESSION PROBLEM Let vectors [x.sub.j] (j = 1, ..., J) be observed independent variables or covariates, and let vectors [[theta].sub.j] be associated latent variables. The data on which the estimation of a binomial regression function...
Bootstrap Confidence Regions for Directional Data
1. INTRODUCTION A common problem in rock magnetism, paleocurrent analysis, and other earth-science areas is calculating a confidence region for the mean direction of a sample of unit vectors, the data typically being circular or spherical. If a...
Characterizations and Analysis of Block Designs
A. K. Nigam, P. D. Puri, and V. K. Gupta. New Delhi: Wiley Eastern, 1988. viii + 176 pp. \$24.95. This book deals with the characterization of block designs in terms of association matrices, concurrence matrices, and efficiency balance. The authors...
Choosing among Alternative Nonexperimental Methods for Estimating the Impact of Social Programs: The Case of Manpower Training
1. INTRODUCTION Evaluation is now an accepted component of most government programs. Federal-government support for evaluating social programs has ranged between 500 million and a billion dollars per year for the past decade. Rossi and Freeman (1985)...
Combinatorial Theory and Statistical Design
Gregory M. Constantine. New York: John Wiley, 1987. xiv + 472 pp. \$52.95. This extremely nice and unusual book gathers interesting, useful combinatorial facts (with proofs) that are normally scattered through books on algebra, graph theory, linear...
Comment
Heckman and Hotz address what they perceive as an imbalance in the research literature on the relative value of experimental and nonexperimental evaluation techniques for estimating the effects of social programs. Specifically, they feel that the literature...
Comment: It's Very Clear
1. INTRODUCTION Nonexperimental studies are here to stay. Heckman and Hotz realize this, and rather than decrying the fact, they make a concerted effort to help improve the interpretation of the results of such studies. They continue a line of research...
Coordinate-Free Multivariable Statistics: An Illustrated Progression from Halmos to Gauss and Bayes
Mervyn Stone. Oxford, U.K.: Clarendon Press, 1987. xiv + 120 pp. \$29.95. In this book, Stone applies a very abstract coordinate-free approach to some results on Gauss-Markov and best-affine Bayes estimators for univariate and multivariate linear...
Empirical Bayes Estimation of Undercount in the Decennial Census
1. INTRODUCTION There are persuasive arguments for and against adjustment of the U.S. decennial census counts, although many of them are based on political rather than technical considerations. The decision whether to adjust depends crucially on...
Empirical Comparison of Approaches to Forming Strata: Using Classification Trees to Adjust for Covariates
1. INTRODUCTION 1.1 The General Problem Medical studies often evaluate the effects of treatments. Outcomes of interest are often affected not only by the treatment, but also by other variables. For example, this article includes data from a recent...
Empirical Model Building
James R. Thompson. New York: John Wiley, 1989. xiv + 242 pp. \$39.95. Textbooks on model building tend to fall into two categories: regression texts that focus on statistical linear models, and applied mathematics texts that focus on differential...
Extreme Value Theory in Engineering
Enrique Castillo. San Diego, CA: Academic Press, 1988. xv + 389 pp. \$59.95. There are very few books on the theory of extreme values of sets of random variables. The first (Gumbel 1958) emphasized applications. More recent are the books by Galambos...
Forecasting Aggregate Period-Specific Birth Rates: The Time Series Properties of a Microdynamic Neoclassical Model of Fertility
1. INTRODUCTION This article demonstrates the value of microdata for understanding the effect of wages on fertility dynamics. Neoclassical economic models of fertility assign a central role to male income and female wages, yet empirical evidence...
Goodness-of-Fit Statistics for Discrete Multivariate Data
Timothy R. C. Read and Noel A. C. Cressie. New York: Springer-Verlag, 1988. xi + 211 pp. \$44. Read and Cressie have written an easy-to-read, informative book that is required reading for anyone analyzing categorical data. They build on their excellent...
How Many Subjects? Statistical Power Analysis in Research
Helena Chmura Kraemer and Sue Thiemann. Newbury Park, CA: Sage Publications, 1987. 120 pp. \$17.95. Power analysis and sample-size calculations have received great attention in recent years: By now, assurances of adequate sample size are de rigueur,...
Introduction to Stochastic Differential Equations
Thomas C. Gard. New York: Marcel Dekker, 1988. xi + 234 pp. \$65 (U.S.A. &amp; Canada); \$78 (elsewhere). Stochastic differential equations (SDE's) are important for characterizing stochastic processes in economics and in management, physical, biological,...
Investigating Smooth Multiple Regression by the Method of Average Derivatives
1. INTRODUCTION The popularity of linear modeling in empirical analysis is based on the ease with which the results can be interpreted. This tradition influenced the modeling of various parametric nonlinear regression relationships, where the mean...
Is Two-Phase Sampling Really Better for Estimating Age Composition?
1. INTRODUCTION Estimates of the age composition of a catch from a fish population is a key ingredient in both catch-curve and virtual population model analyses that aim at estimating the abundance of the stock. Gulland (1977), Seber (1982), Dupont...
Joint Estimation of Incidence and Diagnostic Error Rates from Irregular Longitudinal Data
1. INTRODUCTION Sickle-cell disease, a hereditary disorder of the hemoglobin molecule, is the most common inherited single-gene defect among humans (Gaston et al. 1987). An estimated 50,000 residents of the United States have the disease (Farber,...
Matrix Differential Calculus with Applications in Statistics and Econometrics
Jan R. Magnus and Heinz Neudecker. Chichester, U.K.: John Wiley, 1988. xvii + 393 pp. \$51.95. This book delivers far more than its rather misleading title promises. Statisticians and econometricians who deal with matrix theory in their everyday...
Maximum Entropy and the Lottery
1. INTRODUCTION Each year hundreds of millions of dollars are spent on tickets in government lotteries characterized by an extremely low probability of winning a multimillion dollar jackpot. The most common lottery game is lotto, in which participants...
Median Unbiasedness and Pitman Closeness
1. INTRODUCTION The Pitman closeness (PC) criterion (Pitman 1937) has received a great deal of attention in recent years [namely, Keating (1985); Keating and Mason (1985); Peddada and Khattree (1986); Rao (1981); Rao, Keating, and Mason (1986);...
Methods for Studying Coincidences
... for the 'one chance in a million' will undoubtedly occur, with no less and no more than its appropriate frequency, however surprised we may be that it should occur to us. R. A. Fisher (1937, p. 16) 1. INTRODUCTION Coincidences abound in...
Models for the Analysis of Association in Multivariate Contingency Tables
1. INTRODUCTION Association models for two-way contingency tables frequently provide a more parsimonious and meaningful description of the association in data sets than is typically possible with the usual log-linear models. It is then only natural...
Optimal Design and Analysis of Experiments
Yadolah Dodge, Valeri V. Federov, and Henry P. Wynn (eds.). Amsterdam: North-Holland, 1988. ix + 370 pp. \$105.25. This volume comprises the invited papers for the First International Conference-Workshop on Optimal Design and Analysis of Experiments,...
Optimal Matching for Observational Studies
1. INTRODUCTION 1.1 Two Literatures on Matching There are two essentially disjoint literatures on matching. The first is the statistical literature on the construction of matched samples for observational studies. The second is the literature...
Parameter Estimation in Reliability and Life Span Models
A. Clifford Cohen and Betty Jones Whitten. New York: Marcel Dekker, 1988. xv + 394 pp. \$69.75 (U.S.A. &amp; Canada); \$83.50 (elsewhere). Many excellent books are currently available in the rapidly developing area of reliability theory, survival...
Plane Answers to Complex Questions: The Theory of Linear Models
Ronald Christensen. New York: Springer-Verlag, 1987. xiv + 380 pp. \$42. This well-written and interesting book can serve as a textbook for a graduate-level course in linear model theory and its applications, and as a reference book for a wide range...
Polls and Surveys: Understanding What They Tell Us
Norman M. Bradburn and Seymour Sudman. San Francisco: Jossey-Bass, 1988. xx + 249 pp. \$22.95. When you are next approached at a reception and asked whether this poll is valid or why that person has never been interviewed by a survey organization,...
Probabilistic Measures of Adequacy of a Numerical Search for a Global Maximum
1. INTRODUCTION Iterative methods are necessary to obtain the maximum likelihood estimates of the parameters of many distributions. To succeed, these methods require an optimization algorithm and a vector of starting values of the parameters for...
Probability Distributions on Banach Spaces
N. N. Vakhania, V. I. Tarieladze, and S. A. Chobanyan (W. A. Woyczynski, trans.). Dordrecht, Holland: D. Reidel, 1987. xxvi + 482 pp. \$99. In 1967 Probability Measures on Metric Spaces (Parthasarathy 1967) appeared. In this book the author gave...
Quantitative Risk Assessment for Teratological Effects
1. INTRODUCTION The estimation of risk in developmental toxicity experiments with laboratory animals involves both a qualitative evaluation of whether a compound (e.g., drug or food additive) causes a teratogenic or fetotoxic adverse effect, and...
Regression Analysis of Multivariate Incomplete Failure Time Data by Modeling Marginal Distributions
1. INTRODUCTION The regression analysis of survival data when there is a single and possibly censored event time for each study subject has been extensively investigated [see Cox and Oakes (1984) and Kalbfleisch and Prentice (1980)]. Many studies,...
Rejoinder
1. HOLLAND'S COMMENTS We are grateful to Paul Holland for elevating the level of discourse between econometricians and statisticians working on the common problem of selection bias. It is not surprising in light of the commonalities in the problem...
Risk Analysis of the Space Shuttle: Pre-Challenger Prediction of Failure
1. INTRODUCTION AND SUMMARY On the night of January 27, 1986, the night before the space shuttle Challenger accident, there was a three-hour teleconference among people at Morton Thiokol (manufacturer of the solid rocket motor), Marshall Space Flight...
Robust Statistical Modeling Using the [Dagger] Distribution
1. INTRODUCTION Statistical inference based on the normal distribution (univariate or multivariate) is known to be vulnerable to outliers. Despite this fact and the considerable interest in robust procedures in the mathematical statistical literature,...
Smoothing and Interpolation with the State-Space Model
1. INTRODUCTION This article develops and explores a new result dealing with smoothing for state-space models. Smoothing for state-space models arises in areas such as signal extraction (Kohn and Ansley 1987a), maximum likelihood estimation of parameters...
Some Asymptotic Results for Learning in Single Hidden-Layer Feedforward Network Models
1. INTRODUCTION Cognitive scientists have recently developed a rich and interesting class of nonlinear models inspired by the neural architecture of the brain ("neural network models"), capable of learning through interaction with their environment...
Statistical Methods for the Analysis of Biomedical Data
Robert F. Woolson. New York: John Wiley, 1987. xx + 513 pp. \$44.95. This introduction to biostatistics is intended as a textbook for biomedical students and as a reference for epidemiologic and clinical research investigators. Woolson covers...
Statistical Methods in Cancer Research, Volume 3: The Design and Analysis of Long-Term Animal Experiments (IARC Scientific Publication 79)
J. J. Gart, D. Krewski, P. N. Lee, R. E. Tarone, and J. Wahrendorf (eds.). Lyon, France: International Agency for Research on Cancer, 1986. x + 219 pp. \$45 (available from Oxford University Press). Although books such as Statistical Methods for...
Statistical Methods in Discrimination Litigation
D. H. Kaye and Mikel Aickin (eds.). New York: Marcel Dekker, 1986. x + 218 pp. \$49.75. Kaye and Aickin persuaded six other authors to join them in writing about statistical methods in discrimination cases. The resulting collection of diverse articles...
Statistical Power Analysis for the Behavioral Sciences (2Nd Ed.)
Jacob Cohen. Hillsdale, NJ: Lawrence Erlbaum, 1988. xxi + 567 pp. \$55. This is the second edition of a work that has become a minor classic. It ranks as a required reference for research grant applications in the social sciences, and is largely...
Statistical Reasoning in Law and Public Policy: (Vol. 1) Statistical Concepts and Issues of Fairness; (Vol. 2) Tort Law, Evidence, and Health
Joseph L. Gastwirth. San Diego, CA: Academic Press, 1988. Vol. 1: xxii + 465 pp.; Vol. 2: viii + 458 pp. \$84.50 each. A few years ago, a journalist observed that "what demonstrative evidence was to the 1960s and the early 1970s, statistics has become...
Statistics: A Guide to the Unknown (3Rd Ed.)
Judith M. Tanur, Frederick Mosteller, William H. Kruskal, Erich L. Lehmann, Richard F. Link, Richard S. Pieters, and Gerald R. Rising (eds.). Pacific Grove, CA: Wadsworth &amp; Brooks/Cole, 1989. xxvii + 284 pp. \$16.95. Warren Weaver concludes his...
Statistics for the Life Sciences
Myra L. Samuels. San Francisco: Dellen, 1989. xi + 597 pp. \$38.75. As Samuels's purpose is to introduce statistics to students in the life sciences, the discussion and examples in this textbook emphasize the relevance of statistics to the life sciences....
The Robust Inference for the Cox Proportional Hazards Model
1. INTRODUCTION The Cox proportional hazards model assumes that the hazard function [lambda](t) for the failure time T of an individual with a p vector of covariates Z(t) = ([Z.sub.1](t), ..., [z.sub.p](t))' has the following form: [lambda] (t;...
Unified Theory and Strategies of Survey Sampling
A. Chaudhuri and J. W. E. Vos. Amsterdam: North-Holland, 1988. xvii + 414 pp. \$92.75. This book deals with developments in two major topics in survey sampling: unified theory of sampling in Part A, and sampling with unequal probabilities in Part...