Applied Logistic RegressionWiley, 31. jul. 1989 - 328 sider Shows how to model a binary outcome variable from a linear regression analysis point of view. Develops the logistic regression model and describes its use in methods for modeling the relationship between a dichotomous outcome variable and a set of covariates. Following establishment of the model there is discussion of its interpretation. Several data sets are the source of the examples and the exercises, and a number of software packages are used to analyze data sets, including BMDP, EGRET, GLIM, SAS, and SYSTAT. |
Indhold
Introduction to the Logistic Regression Model | 1 |
Assessment of Fit and Diagnostics for Polytomous | 8 |
Exercises | 58 |
Copyright | |
18 andre sektioner vises ikke
Almindelige termer og sætninger
additional analysis Appendix application approach assess associated assumption calculated Chapter coding computed conditional confounding consider constant continuous controls correct covariate patterns defined degrees of freedom denote dependent described design variables diagnostics dichotomous difference discussed distribution effect equal equation estimated coefficients estimated standard errors examine example expected expression factor Figure follows function important independent variable interaction interpretation interval less levels likelihood ratio test linear regression log-likelihood logistic regression model matched matrix maximum likelihood estimates mean measures method model containing multivariate normal noted observed obtained odds ratio outcome variable p-value package parameters plot possible present probability problem RACE reference residual risk sample scale selection shown significance similar situation slope SMOKE statistics step stratum subjects Table univariate values variance versus weight yields zero