Modelling Binary Data, Second Edition
This book shows how binary data, that is data that can take one of two possible forms, such as alive or dead, and success or failure can be analyzed using statistical modeling. The role of the linear logistic model is particularly stressed, but models based on the complementary log-log transformations are also introduced. Throughout this book, the practical aspects of the modeling approach are emphasized. Indeed the book begins by describing a number of studies in which binary data were recorded.
What people are saying - Write a review
We haven't found any reviews in the usual places.
Statistical inference for binary data
Models for binary and binomial data
Bioassay and some other applications
9 other sections not shown
Other editions - View all
approximate binary data binary observations binomial distribution BMDP case-control study change in deviance chol conﬁdence interval confounding variables constant term constructed variable corresponding data from Example data set deﬁned degrees of freedom deviance on ﬁtting diverticular disease dose ED50 value EGRET equation explanatory variables exposure factor ﬁt ﬁtted model ﬁtted probabilities ﬁtted values ﬁtting the model full model Genstat germination identiﬁed index plot indicator variables individual inﬂuence inﬂuential interaction ith observation likelihood function likelihood residuals linear logistic model link function log acid logdose logistic regression lines logistic regression model logistic transform matched set matrix mean deviance method model ﬁtted model that contains nodal involvement obtained odds ratio outliers overdispersion P-value package parameter estimates particular probit proc probit procedure random effect random variable response probability response variable result rotifers signiﬁcant species standard error standardized deviance residuals statistic success probability variance X2-distribution zero