2
Classical Competing Risks by Martin Crowder Review by: Yanqing Sun Journal of the American Statistical Association, Vol. 97, No. 460 (Dec., 2002), p. 1217 Published by: American Statistical Association Stable URL: http://www.jstor.org/stable/3085872 . Accessed: 15/06/2014 02:01 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Statistical Association. http://www.jstor.org This content downloaded from 91.229.229.162 on Sun, 15 Jun 2014 02:01:51 AM All use subject to JSTOR Terms and Conditions

Classical Competing Risksby Martin Crowder

Embed Size (px)

Citation preview

Page 1: Classical Competing Risksby Martin Crowder

Classical Competing Risks by Martin CrowderReview by: Yanqing SunJournal of the American Statistical Association, Vol. 97, No. 460 (Dec., 2002), p. 1217Published by: American Statistical AssociationStable URL: http://www.jstor.org/stable/3085872 .

Accessed: 15/06/2014 02:01

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to Journalof the American Statistical Association.

http://www.jstor.org

This content downloaded from 91.229.229.162 on Sun, 15 Jun 2014 02:01:51 AMAll use subject to JSTOR Terms and Conditions

Page 2: Classical Competing Risksby Martin Crowder

Book Reviews Book Reviews Book Reviews Book Reviews Book Reviews

Smith, Parke, Balazek, Mitchard, and Krams) and effects of air pollution on childhood respiratory illness (Best, Ickstadt, Wolpert, Cocking, Elliott, Ben- nett, Bottle, and Reed)-are timely and important, and in all cases the statis- tical challenges are significant. The much shorter contributed papers suffer by comparison; the underlying science (largely biomedical) is particularly short- changed.

Alan F. KARR National Institute of Statistical Sciences

Understanding Economic Forecasts.

David F HENDRY and Neil R. ERICSSON (Eds.). Cambridge, MA: MIT Press, 2001. ISBN 0-262-08304-3. ix + 207 pp. $25.95.

"Because of the things [that] we don't know [that] we don't know, the future is largely unpredictable." This quote from Maxine Springer appears on several occasions within this collected set of articles on the art of forecast- ing. Written by eminent economists and practitioners, this volume gives an excellent overview of the main issues underlying economic forecasting. The authors discuss in a balanced way statistical issues as well as economic fun- damentals. I found it most enjoyable and informative to read. The following quote from Shakespeare's Macbeth (I, iii) summarizes our society's longing for a good forecast:

If you can look into the seeds of time And say which grain will grow and which will not, Speak then to me.

No doubt, the contributions in this volume put scientific perspective to this quote.

Paul EMBRECHTS Swiss Federal Institute of Technology

Smith, Parke, Balazek, Mitchard, and Krams) and effects of air pollution on childhood respiratory illness (Best, Ickstadt, Wolpert, Cocking, Elliott, Ben- nett, Bottle, and Reed)-are timely and important, and in all cases the statis- tical challenges are significant. The much shorter contributed papers suffer by comparison; the underlying science (largely biomedical) is particularly short- changed.

Alan F. KARR National Institute of Statistical Sciences

Understanding Economic Forecasts.

David F HENDRY and Neil R. ERICSSON (Eds.). Cambridge, MA: MIT Press, 2001. ISBN 0-262-08304-3. ix + 207 pp. $25.95.

"Because of the things [that] we don't know [that] we don't know, the future is largely unpredictable." This quote from Maxine Springer appears on several occasions within this collected set of articles on the art of forecast- ing. Written by eminent economists and practitioners, this volume gives an excellent overview of the main issues underlying economic forecasting. The authors discuss in a balanced way statistical issues as well as economic fun- damentals. I found it most enjoyable and informative to read. The following quote from Shakespeare's Macbeth (I, iii) summarizes our society's longing for a good forecast:

If you can look into the seeds of time And say which grain will grow and which will not, Speak then to me.

No doubt, the contributions in this volume put scientific perspective to this quote.

Paul EMBRECHTS Swiss Federal Institute of Technology

Smith, Parke, Balazek, Mitchard, and Krams) and effects of air pollution on childhood respiratory illness (Best, Ickstadt, Wolpert, Cocking, Elliott, Ben- nett, Bottle, and Reed)-are timely and important, and in all cases the statis- tical challenges are significant. The much shorter contributed papers suffer by comparison; the underlying science (largely biomedical) is particularly short- changed.

Alan F. KARR National Institute of Statistical Sciences

Understanding Economic Forecasts.

David F HENDRY and Neil R. ERICSSON (Eds.). Cambridge, MA: MIT Press, 2001. ISBN 0-262-08304-3. ix + 207 pp. $25.95.

"Because of the things [that] we don't know [that] we don't know, the future is largely unpredictable." This quote from Maxine Springer appears on several occasions within this collected set of articles on the art of forecast- ing. Written by eminent economists and practitioners, this volume gives an excellent overview of the main issues underlying economic forecasting. The authors discuss in a balanced way statistical issues as well as economic fun- damentals. I found it most enjoyable and informative to read. The following quote from Shakespeare's Macbeth (I, iii) summarizes our society's longing for a good forecast:

If you can look into the seeds of time And say which grain will grow and which will not, Speak then to me.

No doubt, the contributions in this volume put scientific perspective to this quote.

Paul EMBRECHTS Swiss Federal Institute of Technology

Smith, Parke, Balazek, Mitchard, and Krams) and effects of air pollution on childhood respiratory illness (Best, Ickstadt, Wolpert, Cocking, Elliott, Ben- nett, Bottle, and Reed)-are timely and important, and in all cases the statis- tical challenges are significant. The much shorter contributed papers suffer by comparison; the underlying science (largely biomedical) is particularly short- changed.

Alan F. KARR National Institute of Statistical Sciences

Understanding Economic Forecasts.

David F HENDRY and Neil R. ERICSSON (Eds.). Cambridge, MA: MIT Press, 2001. ISBN 0-262-08304-3. ix + 207 pp. $25.95.

"Because of the things [that] we don't know [that] we don't know, the future is largely unpredictable." This quote from Maxine Springer appears on several occasions within this collected set of articles on the art of forecast- ing. Written by eminent economists and practitioners, this volume gives an excellent overview of the main issues underlying economic forecasting. The authors discuss in a balanced way statistical issues as well as economic fun- damentals. I found it most enjoyable and informative to read. The following quote from Shakespeare's Macbeth (I, iii) summarizes our society's longing for a good forecast:

If you can look into the seeds of time And say which grain will grow and which will not, Speak then to me.

No doubt, the contributions in this volume put scientific perspective to this quote.

Paul EMBRECHTS Swiss Federal Institute of Technology

Smith, Parke, Balazek, Mitchard, and Krams) and effects of air pollution on childhood respiratory illness (Best, Ickstadt, Wolpert, Cocking, Elliott, Ben- nett, Bottle, and Reed)-are timely and important, and in all cases the statis- tical challenges are significant. The much shorter contributed papers suffer by comparison; the underlying science (largely biomedical) is particularly short- changed.

Alan F. KARR National Institute of Statistical Sciences

Understanding Economic Forecasts.

David F HENDRY and Neil R. ERICSSON (Eds.). Cambridge, MA: MIT Press, 2001. ISBN 0-262-08304-3. ix + 207 pp. $25.95.

"Because of the things [that] we don't know [that] we don't know, the future is largely unpredictable." This quote from Maxine Springer appears on several occasions within this collected set of articles on the art of forecast- ing. Written by eminent economists and practitioners, this volume gives an excellent overview of the main issues underlying economic forecasting. The authors discuss in a balanced way statistical issues as well as economic fun- damentals. I found it most enjoyable and informative to read. The following quote from Shakespeare's Macbeth (I, iii) summarizes our society's longing for a good forecast:

If you can look into the seeds of time And say which grain will grow and which will not, Speak then to me.

No doubt, the contributions in this volume put scientific perspective to this quote.

Paul EMBRECHTS Swiss Federal Institute of Technology

Le, C. T., and Boen, J. R. (1995), Health & Numbers: Basic Biostatistical Methods, New York: Wiley.

Modeling Intraindividual Variability With Repeated Measures Data.

D. S. MOSKOWITZ and Scott L. HERSHBERGER (Eds.). Mahwah, NJ: Lawrence Erlbaum, 2002. ISBN 0-8058-3125-8. xvi + 276 pp. $59.95.

"This volume began as a nightmare." So begins the Preface of this book (p. ix). According to the authors, the nightmare is the daunting variety of choices that social scientists must make in analyzing longitudinal data. The book is an outgrowth of two symposia presented at the 1997 meeting of the American Psychological Association. There are 9 chapters by a total of 19 authors.

The first two chapters, by Kenny and coworkers and Raudenbush, discuss traditional methods for repeated-measures data, alternative covariance struc- tures, and growth-curve modeling. A chapter by Curran and Hussong presents topics in latent-curve analysis. Ramsay contributes a chapter on functional data analysis. There are two chapters on mixed models, one by Wallace and Green that introduces the topic and one by Singer that shows how to use SAS PROC MIXED to fit individual growth-curve models. Duncan and colleagues write on multilevel modeling, Hilmer discusses time-series regression, and Nesselrode and coworkers contribute a chapter on dynamic factor-analysis models with multivariate time series.

There is evidence that the chapters underwent some review and revision. In addition, the editors have gone to the trouble to compile author and subject indexes and to write a preface that gives a good overview of the book. These features greatly enhance the text's accessibility and usefulness.

Le, C. T., and Boen, J. R. (1995), Health & Numbers: Basic Biostatistical Methods, New York: Wiley.

Modeling Intraindividual Variability With Repeated Measures Data.

D. S. MOSKOWITZ and Scott L. HERSHBERGER (Eds.). Mahwah, NJ: Lawrence Erlbaum, 2002. ISBN 0-8058-3125-8. xvi + 276 pp. $59.95.

"This volume began as a nightmare." So begins the Preface of this book (p. ix). According to the authors, the nightmare is the daunting variety of choices that social scientists must make in analyzing longitudinal data. The book is an outgrowth of two symposia presented at the 1997 meeting of the American Psychological Association. There are 9 chapters by a total of 19 authors.

The first two chapters, by Kenny and coworkers and Raudenbush, discuss traditional methods for repeated-measures data, alternative covariance struc- tures, and growth-curve modeling. A chapter by Curran and Hussong presents topics in latent-curve analysis. Ramsay contributes a chapter on functional data analysis. There are two chapters on mixed models, one by Wallace and Green that introduces the topic and one by Singer that shows how to use SAS PROC MIXED to fit individual growth-curve models. Duncan and colleagues write on multilevel modeling, Hilmer discusses time-series regression, and Nesselrode and coworkers contribute a chapter on dynamic factor-analysis models with multivariate time series.

There is evidence that the chapters underwent some review and revision. In addition, the editors have gone to the trouble to compile author and subject indexes and to write a preface that gives a good overview of the book. These features greatly enhance the text's accessibility and usefulness.

Le, C. T., and Boen, J. R. (1995), Health & Numbers: Basic Biostatistical Methods, New York: Wiley.

Modeling Intraindividual Variability With Repeated Measures Data.

D. S. MOSKOWITZ and Scott L. HERSHBERGER (Eds.). Mahwah, NJ: Lawrence Erlbaum, 2002. ISBN 0-8058-3125-8. xvi + 276 pp. $59.95.

"This volume began as a nightmare." So begins the Preface of this book (p. ix). According to the authors, the nightmare is the daunting variety of choices that social scientists must make in analyzing longitudinal data. The book is an outgrowth of two symposia presented at the 1997 meeting of the American Psychological Association. There are 9 chapters by a total of 19 authors.

The first two chapters, by Kenny and coworkers and Raudenbush, discuss traditional methods for repeated-measures data, alternative covariance struc- tures, and growth-curve modeling. A chapter by Curran and Hussong presents topics in latent-curve analysis. Ramsay contributes a chapter on functional data analysis. There are two chapters on mixed models, one by Wallace and Green that introduces the topic and one by Singer that shows how to use SAS PROC MIXED to fit individual growth-curve models. Duncan and colleagues write on multilevel modeling, Hilmer discusses time-series regression, and Nesselrode and coworkers contribute a chapter on dynamic factor-analysis models with multivariate time series.

There is evidence that the chapters underwent some review and revision. In addition, the editors have gone to the trouble to compile author and subject indexes and to write a preface that gives a good overview of the book. These features greatly enhance the text's accessibility and usefulness.

Le, C. T., and Boen, J. R. (1995), Health & Numbers: Basic Biostatistical Methods, New York: Wiley.

Modeling Intraindividual Variability With Repeated Measures Data.

D. S. MOSKOWITZ and Scott L. HERSHBERGER (Eds.). Mahwah, NJ: Lawrence Erlbaum, 2002. ISBN 0-8058-3125-8. xvi + 276 pp. $59.95.

"This volume began as a nightmare." So begins the Preface of this book (p. ix). According to the authors, the nightmare is the daunting variety of choices that social scientists must make in analyzing longitudinal data. The book is an outgrowth of two symposia presented at the 1997 meeting of the American Psychological Association. There are 9 chapters by a total of 19 authors.

The first two chapters, by Kenny and coworkers and Raudenbush, discuss traditional methods for repeated-measures data, alternative covariance struc- tures, and growth-curve modeling. A chapter by Curran and Hussong presents topics in latent-curve analysis. Ramsay contributes a chapter on functional data analysis. There are two chapters on mixed models, one by Wallace and Green that introduces the topic and one by Singer that shows how to use SAS PROC MIXED to fit individual growth-curve models. Duncan and colleagues write on multilevel modeling, Hilmer discusses time-series regression, and Nesselrode and coworkers contribute a chapter on dynamic factor-analysis models with multivariate time series.

There is evidence that the chapters underwent some review and revision. In addition, the editors have gone to the trouble to compile author and subject indexes and to write a preface that gives a good overview of the book. These features greatly enhance the text's accessibility and usefulness.

Le, C. T., and Boen, J. R. (1995), Health & Numbers: Basic Biostatistical Methods, New York: Wiley.

Modeling Intraindividual Variability With Repeated Measures Data.

D. S. MOSKOWITZ and Scott L. HERSHBERGER (Eds.). Mahwah, NJ: Lawrence Erlbaum, 2002. ISBN 0-8058-3125-8. xvi + 276 pp. $59.95.

"This volume began as a nightmare." So begins the Preface of this book (p. ix). According to the authors, the nightmare is the daunting variety of choices that social scientists must make in analyzing longitudinal data. The book is an outgrowth of two symposia presented at the 1997 meeting of the American Psychological Association. There are 9 chapters by a total of 19 authors.

The first two chapters, by Kenny and coworkers and Raudenbush, discuss traditional methods for repeated-measures data, alternative covariance struc- tures, and growth-curve modeling. A chapter by Curran and Hussong presents topics in latent-curve analysis. Ramsay contributes a chapter on functional data analysis. There are two chapters on mixed models, one by Wallace and Green that introduces the topic and one by Singer that shows how to use SAS PROC MIXED to fit individual growth-curve models. Duncan and colleagues write on multilevel modeling, Hilmer discusses time-series regression, and Nesselrode and coworkers contribute a chapter on dynamic factor-analysis models with multivariate time series.

There is evidence that the chapters underwent some review and revision. In addition, the editors have gone to the trouble to compile author and subject indexes and to write a preface that gives a good overview of the book. These features greatly enhance the text's accessibility and usefulness.

Russell V. LENTH

Health & Numbers: A Problems-Based Introduction The University of Iowa

to Biostatistics (2nd Ed.).

Russell V. LENTH

Health & Numbers: A Problems-Based Introduction The University of Iowa

to Biostatistics (2nd Ed.).

Russell V. LENTH

Health & Numbers: A Problems-Based Introduction The University of Iowa

to Biostatistics (2nd Ed.).

Russell V. LENTH

Health & Numbers: A Problems-Based Introduction The University of Iowa

to Biostatistics (2nd Ed.).

Russell V. LENTH

Health & Numbers: A Problems-Based Introduction The University of Iowa

to Biostatistics (2nd Ed.).

Chap T. LE. New York: Wiley, 2001. ISBN 0-471-41661-4. xiii + 366 pp. $59.95 (P).

This book is a revised and expanded edition of the text of Le and Boen (1995), which was originally reviewed by D'Agostino (1996). The book cov- ers many of the topics typically taught in introductory undergraduate or lower-level graduate courses in statistics for the health sciences. The empha- sis seems to be more on computations and testing than on exploratory data analysis and modeling. For example, the chapter on regression devotes only one full paragraph (on pp. 280-281) and a few remarks made in passing to checks for nonlinearity and nonconstant variance and possible remedies. Graphical diagnostic tools, such as residual plots, are not described (in fact, I do not believe the word "residual" is ever used), and the discussion of model- building techniques is limited to the description of stepwise procedures (the use of which appears to be recommended). The book's main strength is in the numerous datasets pertaining to biostatistical applications, on which the examples and exercises are based. Because of this, the best use of Health & Numbers: A Problems-Based Introduction to Biostatistics might be as a sup- plemental source of illustrations and exercises for an introductory course in biostatistics.

Mario PERUGGIA The Ohio State University

REFERENCES

D'Agostino, R. B. (1996), Review of Health & Numbers: Basic Biostatisti- cal Methods, by C. T. Le and J. R. Boen, The American Statistician, 50, 193-194.

Chap T. LE. New York: Wiley, 2001. ISBN 0-471-41661-4. xiii + 366 pp. $59.95 (P).

This book is a revised and expanded edition of the text of Le and Boen (1995), which was originally reviewed by D'Agostino (1996). The book cov- ers many of the topics typically taught in introductory undergraduate or lower-level graduate courses in statistics for the health sciences. The empha- sis seems to be more on computations and testing than on exploratory data analysis and modeling. For example, the chapter on regression devotes only one full paragraph (on pp. 280-281) and a few remarks made in passing to checks for nonlinearity and nonconstant variance and possible remedies. Graphical diagnostic tools, such as residual plots, are not described (in fact, I do not believe the word "residual" is ever used), and the discussion of model- building techniques is limited to the description of stepwise procedures (the use of which appears to be recommended). The book's main strength is in the numerous datasets pertaining to biostatistical applications, on which the examples and exercises are based. Because of this, the best use of Health & Numbers: A Problems-Based Introduction to Biostatistics might be as a sup- plemental source of illustrations and exercises for an introductory course in biostatistics.

Mario PERUGGIA The Ohio State University

REFERENCES

D'Agostino, R. B. (1996), Review of Health & Numbers: Basic Biostatisti- cal Methods, by C. T. Le and J. R. Boen, The American Statistician, 50, 193-194.

Chap T. LE. New York: Wiley, 2001. ISBN 0-471-41661-4. xiii + 366 pp. $59.95 (P).

This book is a revised and expanded edition of the text of Le and Boen (1995), which was originally reviewed by D'Agostino (1996). The book cov- ers many of the topics typically taught in introductory undergraduate or lower-level graduate courses in statistics for the health sciences. The empha- sis seems to be more on computations and testing than on exploratory data analysis and modeling. For example, the chapter on regression devotes only one full paragraph (on pp. 280-281) and a few remarks made in passing to checks for nonlinearity and nonconstant variance and possible remedies. Graphical diagnostic tools, such as residual plots, are not described (in fact, I do not believe the word "residual" is ever used), and the discussion of model- building techniques is limited to the description of stepwise procedures (the use of which appears to be recommended). The book's main strength is in the numerous datasets pertaining to biostatistical applications, on which the examples and exercises are based. Because of this, the best use of Health & Numbers: A Problems-Based Introduction to Biostatistics might be as a sup- plemental source of illustrations and exercises for an introductory course in biostatistics.

Mario PERUGGIA The Ohio State University

REFERENCES

D'Agostino, R. B. (1996), Review of Health & Numbers: Basic Biostatisti- cal Methods, by C. T. Le and J. R. Boen, The American Statistician, 50, 193-194.

Chap T. LE. New York: Wiley, 2001. ISBN 0-471-41661-4. xiii + 366 pp. $59.95 (P).

This book is a revised and expanded edition of the text of Le and Boen (1995), which was originally reviewed by D'Agostino (1996). The book cov- ers many of the topics typically taught in introductory undergraduate or lower-level graduate courses in statistics for the health sciences. The empha- sis seems to be more on computations and testing than on exploratory data analysis and modeling. For example, the chapter on regression devotes only one full paragraph (on pp. 280-281) and a few remarks made in passing to checks for nonlinearity and nonconstant variance and possible remedies. Graphical diagnostic tools, such as residual plots, are not described (in fact, I do not believe the word "residual" is ever used), and the discussion of model- building techniques is limited to the description of stepwise procedures (the use of which appears to be recommended). The book's main strength is in the numerous datasets pertaining to biostatistical applications, on which the examples and exercises are based. Because of this, the best use of Health & Numbers: A Problems-Based Introduction to Biostatistics might be as a sup- plemental source of illustrations and exercises for an introductory course in biostatistics.

Mario PERUGGIA The Ohio State University

REFERENCES

D'Agostino, R. B. (1996), Review of Health & Numbers: Basic Biostatisti- cal Methods, by C. T. Le and J. R. Boen, The American Statistician, 50, 193-194.

Chap T. LE. New York: Wiley, 2001. ISBN 0-471-41661-4. xiii + 366 pp. $59.95 (P).

This book is a revised and expanded edition of the text of Le and Boen (1995), which was originally reviewed by D'Agostino (1996). The book cov- ers many of the topics typically taught in introductory undergraduate or lower-level graduate courses in statistics for the health sciences. The empha- sis seems to be more on computations and testing than on exploratory data analysis and modeling. For example, the chapter on regression devotes only one full paragraph (on pp. 280-281) and a few remarks made in passing to checks for nonlinearity and nonconstant variance and possible remedies. Graphical diagnostic tools, such as residual plots, are not described (in fact, I do not believe the word "residual" is ever used), and the discussion of model- building techniques is limited to the description of stepwise procedures (the use of which appears to be recommended). The book's main strength is in the numerous datasets pertaining to biostatistical applications, on which the examples and exercises are based. Because of this, the best use of Health & Numbers: A Problems-Based Introduction to Biostatistics might be as a sup- plemental source of illustrations and exercises for an introductory course in biostatistics.

Mario PERUGGIA The Ohio State University

REFERENCES

D'Agostino, R. B. (1996), Review of Health & Numbers: Basic Biostatisti- cal Methods, by C. T. Le and J. R. Boen, The American Statistician, 50, 193-194.

Classical Competing Risks.

Martin CROWDER. New York: Chapman and Hall/CRC, 2001. ISBN 1-58488-175-5. xiv+ 186 pp. $74.95.

This informative book provides a detailed account of competing risks, including the associated likelihood functions and the hazard-based approach. The book includes eight chapters. The basic quantities for competing risks and various models are described in Chapter 1, and parametric likelihood inference is covered in Chapter 2. Chapter 3 describes the approach to com- peting risks via the joint distribution of latent failure times. Chapter 4 covers survival analysis via hazard functions. Chapters 5 and 6 carry this work to discrete and continuous failure times in competing risks. Chapter 7 discusses the issues that arise with the latent failure times approach. The last chapter gives a brief introduction to the counting-process approach.

Numerous data examples and a rich class of statistical models are intro- duced and analyzed. The statistical theorems are presented with very clear background and references. The book presents parametric and nonparam- etric methods for both discrete and continuous failure times. The important issues are thoroughly discussed, including marginal versus subdistributions and latent failure times versus sub-hazard function approaches, are discussed in depth. It seems that the reference to Press et al. (1990) on page 170 is missing from the bibliography. Otherwise, Classical Competing Risks is self- contained and well written at a level accessible to graduate students and applied statisticians alike.

Yanqing SUN University of North Carolina Charlotte

Classical Competing Risks.

Martin CROWDER. New York: Chapman and Hall/CRC, 2001. ISBN 1-58488-175-5. xiv+ 186 pp. $74.95.

This informative book provides a detailed account of competing risks, including the associated likelihood functions and the hazard-based approach. The book includes eight chapters. The basic quantities for competing risks and various models are described in Chapter 1, and parametric likelihood inference is covered in Chapter 2. Chapter 3 describes the approach to com- peting risks via the joint distribution of latent failure times. Chapter 4 covers survival analysis via hazard functions. Chapters 5 and 6 carry this work to discrete and continuous failure times in competing risks. Chapter 7 discusses the issues that arise with the latent failure times approach. The last chapter gives a brief introduction to the counting-process approach.

Numerous data examples and a rich class of statistical models are intro- duced and analyzed. The statistical theorems are presented with very clear background and references. The book presents parametric and nonparam- etric methods for both discrete and continuous failure times. The important issues are thoroughly discussed, including marginal versus subdistributions and latent failure times versus sub-hazard function approaches, are discussed in depth. It seems that the reference to Press et al. (1990) on page 170 is missing from the bibliography. Otherwise, Classical Competing Risks is self- contained and well written at a level accessible to graduate students and applied statisticians alike.

Yanqing SUN University of North Carolina Charlotte

Classical Competing Risks.

Martin CROWDER. New York: Chapman and Hall/CRC, 2001. ISBN 1-58488-175-5. xiv+ 186 pp. $74.95.

This informative book provides a detailed account of competing risks, including the associated likelihood functions and the hazard-based approach. The book includes eight chapters. The basic quantities for competing risks and various models are described in Chapter 1, and parametric likelihood inference is covered in Chapter 2. Chapter 3 describes the approach to com- peting risks via the joint distribution of latent failure times. Chapter 4 covers survival analysis via hazard functions. Chapters 5 and 6 carry this work to discrete and continuous failure times in competing risks. Chapter 7 discusses the issues that arise with the latent failure times approach. The last chapter gives a brief introduction to the counting-process approach.

Numerous data examples and a rich class of statistical models are intro- duced and analyzed. The statistical theorems are presented with very clear background and references. The book presents parametric and nonparam- etric methods for both discrete and continuous failure times. The important issues are thoroughly discussed, including marginal versus subdistributions and latent failure times versus sub-hazard function approaches, are discussed in depth. It seems that the reference to Press et al. (1990) on page 170 is missing from the bibliography. Otherwise, Classical Competing Risks is self- contained and well written at a level accessible to graduate students and applied statisticians alike.

Yanqing SUN University of North Carolina Charlotte

Classical Competing Risks.

Martin CROWDER. New York: Chapman and Hall/CRC, 2001. ISBN 1-58488-175-5. xiv+ 186 pp. $74.95.

This informative book provides a detailed account of competing risks, including the associated likelihood functions and the hazard-based approach. The book includes eight chapters. The basic quantities for competing risks and various models are described in Chapter 1, and parametric likelihood inference is covered in Chapter 2. Chapter 3 describes the approach to com- peting risks via the joint distribution of latent failure times. Chapter 4 covers survival analysis via hazard functions. Chapters 5 and 6 carry this work to discrete and continuous failure times in competing risks. Chapter 7 discusses the issues that arise with the latent failure times approach. The last chapter gives a brief introduction to the counting-process approach.

Numerous data examples and a rich class of statistical models are intro- duced and analyzed. The statistical theorems are presented with very clear background and references. The book presents parametric and nonparam- etric methods for both discrete and continuous failure times. The important issues are thoroughly discussed, including marginal versus subdistributions and latent failure times versus sub-hazard function approaches, are discussed in depth. It seems that the reference to Press et al. (1990) on page 170 is missing from the bibliography. Otherwise, Classical Competing Risks is self- contained and well written at a level accessible to graduate students and applied statisticians alike.

Yanqing SUN University of North Carolina Charlotte

Classical Competing Risks.

Martin CROWDER. New York: Chapman and Hall/CRC, 2001. ISBN 1-58488-175-5. xiv+ 186 pp. $74.95.

This informative book provides a detailed account of competing risks, including the associated likelihood functions and the hazard-based approach. The book includes eight chapters. The basic quantities for competing risks and various models are described in Chapter 1, and parametric likelihood inference is covered in Chapter 2. Chapter 3 describes the approach to com- peting risks via the joint distribution of latent failure times. Chapter 4 covers survival analysis via hazard functions. Chapters 5 and 6 carry this work to discrete and continuous failure times in competing risks. Chapter 7 discusses the issues that arise with the latent failure times approach. The last chapter gives a brief introduction to the counting-process approach.

Numerous data examples and a rich class of statistical models are intro- duced and analyzed. The statistical theorems are presented with very clear background and references. The book presents parametric and nonparam- etric methods for both discrete and continuous failure times. The important issues are thoroughly discussed, including marginal versus subdistributions and latent failure times versus sub-hazard function approaches, are discussed in depth. It seems that the reference to Press et al. (1990) on page 170 is missing from the bibliography. Otherwise, Classical Competing Risks is self- contained and well written at a level accessible to graduate students and applied statisticians alike.

Yanqing SUN University of North Carolina Charlotte

1217 1217 1217 1217 1217

This content downloaded from 91.229.229.162 on Sun, 15 Jun 2014 02:01:51 AMAll use subject to JSTOR Terms and Conditions