31. |
Diagnostics for Linearization Confidence Intervals in Nonlinear Regression |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1068-1074
Jian-Shen Chen,
RobertI. Jennrich,
Preview
|
PDF (706KB)
|
|
摘要:
We investigate linear approximation (LA) confidence intervals for functionsg(θ) of the parameters θ in a nonlinear regression model. These intervals are almost universally used and generally perform well, but at times they have poor coverage probabilities. A diagnostic plot and index are developed to detect these failures. We show how these diagnostics may be used to estimate coverage probabilities and these are used to calibrate the diagnostics. The performance of the coverage probability estimates in a variety of nonlinear regression problems is investigated via simulation; for these problems, they work quite well. Conditions are identified under which the estimates are exact. Finally, we discuss the use of the profiletplot and asymmetry and bias indices as diagnostics for LA intervals and show how to calibrate them in terms of coverage probabilities.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476609
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
32. |
Estimators of Odds Ratio Regression Parameters in Matched Case-Control Studies with Covariate Measurement Error |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1075-1084
AndrewB. Forbes,
ThomasJ. Santner,
Preview
|
PDF (1060KB)
|
|
摘要:
This article studies estimators of the odds ratio and odds ratio regression parameters in finely matched case-control studies containing a binary exposure of primary interest and subject-specific covariates that are subject to measurement error. A retrospective logistic regression model for the binary exposure variable is used. The effect of measurement errors on the conditional maximum likelihood estimator is determined. Three alternatives are considered: bias-corrected, functional, and “transformation” estimators. The asymptotic and small-sample properties of the three competitors are studied. The results are illustrated using data from a case-control study of diet and colon cancer.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476610
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
33. |
Comparison of Regression Curves Using Quasi-Residuals |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1085-1093
K.B. Kulasekera,
Preview
|
PDF (862KB)
|
|
摘要:
We consider testing the equality of two regression functions using two independent samples. Three tests are proposed that are free of the restriction of having the same covariate values or sample sizes for both samples. Asymptotic distributions are given and results from a simulation study are presented that show the superior power properties of these tests over a competing test in a variety of cases, including the testing of hypotheses involving high-frequency curves when the design points for the two samples differ. It is also observed that the tests have good level properties when proper smoothing parameters are selected.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476611
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
34. |
Simultaneous Confidence Bands for Linear Regression with Heteroscedastic Errors |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1094-1098
JulianJ. Faraway,
Jiayang Sun,
Preview
|
PDF (435KB)
|
|
摘要:
The Scheffé method may be used to construct simultaneous confidence bands for a regression surface for the whole predictor space. When the bands need only hold for a subset of that space, previous authors have described how the bands can be appropriately narrowed while still maintaining the desired level of confidence. Data with heteroscedastic errors occur often, and unless some transformation is feasible, there is no obvious way to construct bands using the current methods. This article shows how to construct approximate simultaneous confidence bands when the errors are heteroscedastic and symmetric. The method works when the weights are known or unknown and have to be estimated. The region in which the bands must hold can be quite general and will work for any linear unbiased estimate of the regression surface. The method can even be extended to linear estimates with a small amount of bias such as nonparametric kernel regression smoothers.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476612
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
35. |
An Alternative Definition of Finite-Sample Breakdown Point with Applications to Regression Model Estimators |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1099-1106
Shinichi Sakata,
Halbert White,
Preview
|
PDF (805KB)
|
|
摘要:
We propose an alternative definition of the finite-sample breakdown point. This breakdown point is invariant with respect to reparameterization and compatible with the Donoho-Huber breakdown point in linear regression situations. It also overcomes certain limitations of the definition proposed by Stromberg and Ruppert and can be used in a wide range of estimation problems. We investigate the breakdown properties of some nonlinear regression estimators. These results alert us to the danger of using familiar M estimators with data sets containing outliers and to the advantages of using estimators based on Hampel's proposal, such as S estimators.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476613
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
36. |
Minimax Estimation of Proportions under Random Sample Size |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1107-1111
Peter Amrhein,
Preview
|
PDF (388KB)
|
|
摘要:
We study the problem of how to estimate the parameters of a multivariate hypergeometric distribution when the sample sizenis assumed to be an ancillary statistic. The minimax estimator for squared error loss is given. This estimator differs from the well-known minimax estimator for fixedn. Furthermore, that classical estimator is shown to be not even admissible whennis random.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476614
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
37. |
Modeling the Drop-Out Mechanism in Repeated-Measures Studies |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1112-1121
RoderickJ. A. Little,
Preview
|
PDF (1185KB)
|
|
摘要:
Subjects often drop out of longitudinal studies prematurely, yielding unbalanced data with unequal numbers of measures for each subject. Modern software programs for handling unbalanced longitudinal data improve on methods that discard the incomplete cases by including all the data, but also yield biased inferences under plausible models for the drop-out process. This article discusses methods that simultaneously model the data and the drop-out process within a unified model-based framework. Models are classified into two broad classes—random-coefficient selection models and random-coefficient pattern-mixture models—depending on how the joint distribution of the data and drop-out mechanism is factored. Inference is likelihood-based, via maximum likelihood or Bayesian methods. A number of examples in the literature are placed in this framework, and possible extensions outlined. Data collection on the nature of the drop-out process is advocated to guide the choice of model. In cases where the drop-out mechanism is not well understood, sensitivity analyses are suggested to assess the effect on inferences about target quantities of alternative assumptions about the drop-out process.
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476615
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
38. |
Book Reviews |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1122-1135
Preview
|
PDF (2304KB)
|
|
摘要:
Analysis of Longitudinal Data.Peter J. Diggle, Kung-Yee Liang, and Scott L. Zeger. New York: Oxford University Press, 1994. xi + 253 pp. $36. Reviewed by Thomas R. Ten HaveThe Pennsylvania State University
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476616
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
39. |
Telegraphic Reviews |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1135-1136
Preview
|
PDF (280KB)
|
|
摘要:
Applied Econometric Time Series.Walter Enders. New York: John Wiley, 1995. xi + 433 pp. $55.95. Reviewed by MW
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476617
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|
40. |
Correction |
|
Journal of the American Statistical Association,
Volume 90,
Issue 431,
1995,
Page 1136-1136
Preview
|
PDF (135KB)
|
|
ISSN:0162-1459
DOI:10.1080/01621459.1995.10476618
出版商:Taylor & Francis Group
年代:1995
数据来源: Taylor
|