Access the full text.
Sign up today, get DeepDyve free for 14 days.
Hanley Hanley, McNeil McNeil (1982)
The meaning and use of the area under a receiver operating characteristic (ROC) curveRadiology, 143
Roecker Roecker (1991)
Prediction error and its estimation for subset‐selected modelsTechnometrics, 33
Cox Cox (1958)
The regression analysis of binary sequences (with discussion)Journal of the Royal Statistical Society, Series B, 20
Landwehr Landwehr, Pregibon Pregibon, Shoemaker Shoemaker (1984)
Graphical methods for assessing logistic regression models (with discussion)Journal of the American Statistical Association, 79
Mantel Mantel (1970)
Why stepdown procedures in variable selectionTechnometrics, 12
Crawford Crawford, Tennstedt Tennstedt, McKinlay McKinlay (1995)
A comparison of analytic methods for nonrandom missingness of outcome dataJournal of Clinical Epidemiology, 48
Marshall Marshall, Grover Grover, Henderson Henderson, Hammermeister Hammermeister (1994)
Assessment of predictive models for binary outcomes: an empirical approach using operative death from cardiac surgeryStatistics in Medicine, 13
Sauerbrei Sauerbrei, Schumacher Schumacher (1992)
A bootstrap resampling procedure for model building: Application to the Cox regression modelStatistics in Medicine, 11
Bamber Bamber (1975)
The area above the ordinal dominance graph and the area below the receiver operating characteristic graphJournal of Mathematical Psychology, 12
van Houwelingen van Houwelingen, le Cessie le Cessie (1988)
Logistic regression, a reviewStatistica Neerlandica, 42
Walker Walker, Duncan Duncan (1967)
Estimation of the probability of an event as a function of several independent variablesBiometrika, 54
Durrleman Durrleman, Simon Simon (1989)
Flexible regression models with cubic splinesStatistics in Medicine, 8
Sleeper Sleeper, Harrington Harrington (1990)
Regression splines in the Cox model with application to covariate effects in liver diseaseJournal of the American Statistical Association, 85
Schemper Schemper (1984)
Analyses of associations with censored data by generalized Mantel and Breslow tests and generalized Kendall correlationBiometrical Journal, 26
Hurvich Hurvich, Tsai Tsai (1990)
The impact of model selection on inference in linear regressionAmerican Statistician, 44
Pettitt Pettitt, Bin Daud Bin Daud (1990)
Investigating time dependence in Cox's proportional hazards modelApplied Statistics, 39
Knaus Knaus, Harrell Harrell, Fisher Fisher, Wagner Wagner, Opan Opan, Sadoff Sadoff, Draper Draper, Walawander Walawander, Conboy Conboy, Grasela Grasela (1993)
The clinical evaluation of new drugs for sepsis: A prospective study design based on survival analysisJournal of the American Medical Association, 270
Liu Liu, Dyer Dyer (1979)
A rank statistic for assessing the amount of variation explained by risk factors in epidemiologic studiesAmerican Journal of Epidemiology, 109
Harrell Harrell, Califf Califf, Pryor Pryor, Lee Lee, Rosati Rosati (1982)
Evaluating the yield of medical testsJournal of the American Medical Association, 247
Harrell Harrell, Lee Lee, Pollock Pollock (1988)
Regression models in clinical studies: determining relationships between predictors and responseJournal of the National Cancer Institute, 80
Efron Efron (1986)
How biased is the apparent error rate of a prediction rule?Journal of the American Statistical Association, 81
Cleveland Cleveland (1979)
Robust locally weighted regression and smoothing scatterplotsJournal of the American Statistical Association, 74
Schoenfeld Schoenfeld (1982)
Partial residuals for the proportional hazards regression modelBiometrika, 69
Nagelkerke Nagelkerke (1991)
A note on a general definition of the coefficient of determinationBiometrika, 78
Breiman Breiman (1992)
The little bootstrap and other methods for dimensionality selection in regression: X‐fixed prediction errorJournal of the American Statistical Association, 87
Therneau Therneau, Grambsch Grambsch, Fleming Fleming (1990)
Martingale‐based residuals for survival modelsBiometrika, 77
Altman Altman, Andersen Andersen (1989)
Bootstrap investigation of the stability of a Cox regression modelStatistics in Medicine, 8
Schemper Schemper (1988)
Non‐parametric analysis of treatment‐covariate interaction in the presence of censoringStatistics in Medicine, 7
Byar Byar, Green Green (1980)
The choice of treatment for cancer patients based on covariate information: application to prostate cancerBulletin Cancer, Paris, 67
Efron Efron, Gong Gong (1983)
A leisurely look at the bootstrap, the jackknife, and cross‐validationAmerican Statistician, 37
Timm Timm (1970)
The estimation of variance‐covariance and correlation matrices from incomplete dataPsychometrika, 35
Verweij Verweij, van Houwelingen van Houwelingen (1994)
Penalized likelihood in Cox regressionStatistics in Medicine, 13
Copas Copas (1983)
Regression, prediction and shrinkage (with discussion)Journal of the Royal Statistical Society, Series B, 45
Linnet Linnet (1989)
Assessing diagnostic tests by a strictly proper scoring ruleStatistics in Medicine, 8
Atkinson Atkinson (1980)
A note on the generalized information criterion for choice of a modelBiometrika, 67
Efron Efron (1983)
Estimating the error rate of a prediction rule: improvement on cross‐validationJournal of the American Statistical Association, 78
Buck Buck (1960)
A method of estimation of missing values in multivariate data suitable for use with an electronic computerJournal of the Royal Statistical Society, Series B, 22
Donner Donner (1982)
The relative effectiveness of procedures commonly used in multiple regression analysis for dealing with missing valuesAmerican Statistician, 36
van Houwelingen van Houwelingen, le Cessie le Cessie (1990)
Predictive value of statistical modelsStatistics in Medicine, 8
Copas Copas (1987)
Cross‐validation shrinkage of regression predictorsJournal of the Royal Statistical Society, Series B, 49
Kaplan Kaplan, Meier Meier (1958)
Nonparametric estimation from incomplete observationsJournal of the American Statistical Association, 53
Cox Cox (1972)
Regression models and life‐tables (with discussion)Journal of the Royal Statistical Society, Series B, 34
Grambsch Grambsch, O'Brien O'Brien (1991)
The effects of transformations and preliminary tests for non‐linearity in regressionStatistics in Medicine, 10
Gray Gray (1992)
Flexible methods for analyzing survival data using splines, with applications to breast cancer prognosisJournal of the American Statistical Association, 87
Kay Kay (1986)
Treatment effects in competing‐risks analysis of prostate cancer dataBiometrics, 42
Picard Picard, Berk Berk (1990)
Data splittingAmerican Statistician, 44
Lee Lee, Pryor Pryor, Harrell Harrell, Califf Califf, Behar Behar, Floyd Floyd, Morris Morris, Waugh Waugh, Whalen Whalen, Rosati Rosati (1986)
Predicting outcome in coronary disease: Statistical models versus expert cliniciansAmerican Journal of Medicine, 80
Grambsch Grambsch, Therneau Therneau (1994)
Proportional hazards tests and diagnostics based on weighted residualsBiometrika, 81
Spiegelhalter Spiegelhalter (1986)
Probabilistic prediction in patient managementStatistics in Medicine, 5
Brier Brier (1950)
Verification of forecasts expressed in terms of probabilityMonthly Weather Review, 75
Korn Korn, Simon Simon (1990)
Measures of explained variation for survival dataStatistics in Medicine, 9
Derksen Derksen, Keselman Keselman (1992)
Backward, forward and stepwise automated subset selection algorithms: Frequency of obtaining authentic and noise variablesBritish Journal of Mathematical and Statistical Psychology, 45
Harrell Harrell, Lee Lee, Califf Califf, Pryor Pryor, Rosati Rosati (1984)
Regression modelling strategies for improved prognostic predictionStatistics in Medicine, 3
Multivariable regression models are powerful tools that are used frequently in studies of clinical outcomes. These models can use a mixture of categorical and continuous variables and can handle partially observed (censored) responses. However, uncritical application of modelling techniques can result in models that poorly fit the dataset at hand, or, even more likely, inaccurately predict outcomes on new subjects. One must know how to measure qualities of a model's fit in order to avoid poorly fitted or overfitted models. Measurement of predictive accuracy can be difficult for survival time data in the presence of censoring. We discuss an easily interpretable index of predictive discrimination as well as methods for assessing calibration of predicted survival probabilities. Both types of predictive accuracy should be unbiasedly validated using bootstrapping or cross‐validation, before using predictions in a new data series. We discuss some of the hazards of poorly fitted and overfitted regression models and present one modelling strategy that avoids many of the problems discussed. The methods described are applicable to all regression models, but are particularly needed for binary, ordinal, and time‐to‐event outcomes. Methods are illustrated with a survival analysis in prostate cancer using Cox regression.
Statistics in Medicine – Wiley
Published: Feb 29, 1996
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.