Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models

摘要: This new book provides a unified, in-depth, readable introduction to the multipredictor regression methods most widely used in biostatistics: linear models for continuous outcomes, logistic binary Cox model right-censored survival times, repeated-measures longitudinal and hierarchical generalized counts other outcomes. Treating these topics together takes advantage of all they have common. The authors point out many-shared elements present selecting, estimating, checking, interpreting each models. They also show that deal with confounding, mediation, interaction causal effects essentially same way. The examples, analyzed using Stata, are drawn from biomedical context but generalize areas application. While first course statistics is assumed, chapter reviewing basic statistical included. Some advanced covered presentation remains intuitive. A brief analysis complex surveys notes further reading provided. For many students researchers learning use methods, this one may be need conduct interpret analyses. The on faculty Division Biostatistics, Department Epidemiology University California, San Francisco, or co-authors more than 200 methodological as well applied papers biological sciences. senior author, Charles E. McCulloch, head author Generalized Linear Mixed Models (2003), Generalized, Linear, (2000), Variance Components (1992). From reviews: "This unified listed title...The illustrated by data medical studies...A real strength careful discussion issues common covered." Journal Biopharmaceutical Statistics, 2005 "This not just biostatisticians. It is, fact, very good, relatively nonmathematical, overview Although examples biologically oriented, generally easy understand follow...I heartily recommend book" Technometrics, February 2006 "Overall, text an particularly strong its breadth coverage emphasis insight place mathematical detail. As intended, well-unified approach should appeal who learn conceptually verbally." American Statistical Association, March 2006