You are previewing Analyzing Receiver Operating Characteristic Curves with SAS.
O'Reilly logo
Analyzing Receiver Operating Characteristic Curves with SAS

Book Description

As a diagnostic decision-making tool, receiver operating characteristic (ROC) curves provide a comprehensive and visually attractive way to summarize the accuracy of predictions. They are used extensively in medical diagnosis and increasingly in fields such as data mining, credit scoring, weather forecasting, and psychometry. In Analyzing Receiver Operating Characteristic Curves with SAS, author Mithat Gonen illustrates the many existing SAS procedures that can be tailored to produce ROC curves and expands upon further analyses using other SAS procedures and macros. Both parametric and nonparametric methods for analyzing ROC curves are covered in detail.

Topics addressed include:

  • Appropriate methods for binary, ordinal, and continuous measures
  • Computations using PROC FREQ, PROC LOGISTIC, PROC NLMIXED, and macros
  • Comparing the ROC curves of several markers and adjusting them for covariates
  • ROC curves with censored data
  • Using the ROC curve for evaluating multivariable prediction models via bootstrap and cross-validation
  • ROC curves in SAS Enterprise Miner
  • And more!

Written for any statistician interested in learning more about ROC curve methodology, the book assumes readers have a basic understanding of regression procedures and moderate familiarity with Base SAS and SAS/STAT. Some familiarity with SAS/GRAPH is helpful but not essential.

This book is part of the SAS Press program.

Table of Contents

  1. Cover Page
  2. Title Page
  3. Copyright Page
  4. Contents
  5. Preface
  6. Acknowledgments
  7. Chapter 1 - Introduction
    1. 1.1 - About Receiver Operating Characteristic Curves
    2. 1.2 - Summary of Chapters
  8. Chapter 2 - Single Binary Predictor
    1. 2.1 - Introduction
    2. 2.2 - Frost Forecast Example
    3. 2.3 - Misclassification Rate
    4. 2.4 - Sensitivity and Specificity
    5. 2.5 - Computations Using PROC FREQ
  9. Chapter 3 - Single Continuous Predictor
    1. 3.1 - Dichotomizing a Continuous Predictor
    2. 3.2 - The ROC Curve
    3. 3.3 - Empirical ROC Curve and the Conditional Distributions of the Marker
    4. 3.4 - Area under the ROC Curve
    5. 3.5 - Selecting an Optimal Threshold
    6. 3.6 - The Binormal ROC Curve
    7. 3.7 - Transformations to Binormality
    8. 3.8 - Direct Estimation of the Binormal ROC Curve
    9. 3.9 - Bootstrap Confidence Intervals for the Area Under the Curve
  10. Chapter 4 - Comparison and Covariate Adjustment of ROC Curves
    1. 4.1 - Introduction
    2. 4.2 - An Example from Prostate Cancer Prognosis
    3. 4.3 - Paired Versus Unpaired Comparisons
    4. 4.4 - Comparing the Areas Under the Empirical ROC Curves
    5. 4.5 - Comparing the Binormal ROC Curves
    6. 4.6 - Discrepancy Between Binormal and Empirical ROC Curves
    7. 4.7 - Bootstrap Confidence Intervals for the Difference in the Area Under the Empirical ROC Curve
    8. 4.8 - Covariate Adjustment for ROC Curves
    9. 4.9 - Regression Model for the Binormal ROC Curve
  11. Chapter 5 - Ordinal Predictors
    1. 5.1 - Introduction
    2. 5.2 - Credit Rating Example
    3. 5.3 - Empirical ROC Curve for Ordinal Predictors
    4. 5.4 - Area Under the Empirical ROC Curve
    5. 5.5 - Latent Variable Model
    6. 5.6 - Comparing ROC Curves for Ordinal Markers
  12. Chapter 6 - Lehmann Family of ROC Curves
    1. 6.1 - Introduction
    2. 6.2 - Lehmann Family of Distributions
    3. 6.3 - Magnetic Resonance Example
    4. 6.4 - Adjusting for Covariates
    5. 6.5 - Using Estimating Equations to Handle Clustered Data
    6. 6.6 - Comparing Markers Using the Lehmann Family of ROC Curves
    7. 6.7 - Advantages and Disadvantages of the Lehmann Family of ROC Curves
  13. Chapter 7 - ROC Curves with Censored Data
    1. 7.1 - Introduction
    2. 7.2 - Lung Cancer Example
    3. 7.3 - ROC Curves with Censored Data
    4. 7.4 - Concordance Probability with Censored Data
    5. 7.5 - Concordance Probability and the Cox Model
  14. Chapter 8 - Using the ROC Curve to Evaluate Multivariable Prediction Models
    1. 8.1 - Introduction
    2. 8.2 - Liver Surgery Example
    3. 8.3 - Resubstitution Estimate of the ROC Curve
    4. 8.4 - Split-Sample Estimates of the ROC Curve
    5. 8.5 - Cross-Validation Estimates of the ROC Curve
    6. 8.6 - Bootstrap-Validated Estimates of the ROC Curve
  15. Chapter 9 - ROC Curves in SAS Enterprise Miner
    1. 9.1 - Introduction
    2. 9.2 - Home Equity Loan Example
    3. 9.3 - ROC Curves from SAS Enterprise Miner for a Single Model
    4. 9.4 - ROC Curves from SAS Enterprise Miner for Competing Models
    5. 9.5 - ROC Curves Using PROC GPLOT with Exported Data from SAS Enterprise Miner
  16. Appendix An Introduction to PROC NLMIXED
    1. A.1 - Fitting a Simple Linear Model: PROC GLM vs PROC NLMIXED
    2. A.2 - PROC NLMIXED and the Binormal Model
  17. References
  18. Index
  19. Accelerate Your SAS Knowledge with SAS Books