azheart: azheart

In LOGIT: Functions, Data and Code for Binary and Binomial Data

Description

Random subset of 34 patients from the 1991 Arizona Medicare data for patients hospitalized subsequent to undergoing a CABG (DRGs 106, 107) or PTCA (DRG 112) cardiovascular procedure.

Usage

data(azheart)

Format

A data frame with 34 observations on the following 6 variables.

died

1=died as a result of surgery; 0=not died

procedure

1=CABG; 0=PTCA

age

age of subject

gender

1=Male; 0=Female

los

hospital length of stay

type

1=emerg/urgent admission; 0=elective admission

Details

azheart is saved as a data frame.

Source

Hilbe, Practical Guide to Logistic Regression, Chapman & Hall/CRC

References

Hilbe, Joseph M (2015), Practical Guide to Logistic Regression, Chapman & Hall/CRC

Examples

library(LOGIT)
#library(COUNT)
data(azheart); attach(azheart)
table(los); table(procedure, type); table(los, died)
summary(los)
summary(mymod <- glm(died ~ procedure + type + los, family=binomial, data=azheart))
#modelfit(mymod)
summary(mymodq <- glm(died ~ procedure+ type  + los, family=quasibinomial, data=azheart))
#modelfit(mymodq)
#library(sandwich)
#sqrt(diag(vcovHC(mymod, type="HC0")))
toOR(mymod)

LOGIT documentation built on May 29, 2017, 10:26 a.m.