作者:
出版社:
简介:Summary:
Publisher Summary 1
Logistic regression is a generalized linear model used for binomial regression for the purposes of predicting the probability of occurrence of an event by fitting data to a logistic curve. In this text, Menard (Sam Houston State U.) goes beyond his earlier monograph, Applied Logistic Regression Analysis (Sage, 2002), by offering more detailed discussion of technical details and practical applications. The text introduces logistic regression analysis from the perspectives of ordinary least squares linear regression analysis and log-linear analysis, deals with global model statistics, discusses interpretation of coefficients and inferential statistics for individual predictors in logistic regression, examines the diagnosis of and remedies for the problems in the dichotomous logistic regression model, details how path analysis can be applied in the logistic regression framework with categorical dependent and intervening variables as well as predictors, deals with logistic regression for polytomous dependent variables, discusses adjustments to the logistic regression model when data involve clustered or other dependent samples (including contextual dependencies), describes the use of logistic regression analysis for longitudinal data with few and many repeated measurements, examines multilevel change models and event history analysis using the proportional odds logistic regression model to model the relationship between categorical dependent variables and time dimensions, and briefly compares logistic regression to other methods of analysis such as probit models and discriminant analysis. A background in basic statistics through ordinary least squares regression is the minimal prerequisite, but background in multiple regression analysis and log-linear analysis will aid in understanding. Annotation 漏2009 Book News, Inc., Portland, OR (booknews.com)
Publisher Summary 2
In this text, author Scott Menard provides coverage of not only the basic logistic regression model but also advanced topics found in no other logistic regression text. The book keeps mathematical notation to a minimum, making it accessible to those with more limited statistics backgrounds, while including advanced topics of interest to more statistically sophisticated readers. Not dependent on any one software package, the book discusses limitations to existing software packages and ways to overcome them.
Key Features聽聽聽Examines the logistic regression model in detailIllustrates concepts with applied examples to help readers understand how concepts are translated into the logistic regression model聽Helps readers make decisions about the criteria for evaluating logistic regression models through detailed coverage of how to assess overall models and individual predictors for categorical dependent variables聽Offers unique coverage of path analysis with logistic regression that shows readers how to examine both direct and indirect effects using logistic regression analysis聽Applies logistic regression analysis to longitudinal panel data, helping students understand the issues in measuring change with dichotomous, nominal, and ordinal dependent variablesShows readers how multilevel change models with logistic regression are different from multilevel growth curve models for continuous interval or ratio-scaled dependent variablesLogistic Regression is intended for courses such as Regression and Correlation, Intermediate/Advanced Statistics, and Quantitative Methods taught in departments throughout the behavioral, health, mathematical, and social sciences, including applied mathematics/statistics, biostatistics, criminology/criminal justice, education, political science, public health/epidemiology, psychology, and sociology.