Logistic Regression, binned residual plot, and calibration plot

General
1

Hi, I am stuck with an assignment question:
Fit a logistic regression model to predict 10 year probability of heart disease given cigarettes per day, sex, age, and education. Summarize the association between smoking and heart disease using your model. For example, report the difference in predicted probability of heart disease in smokers and non-smokers at several levels of sex, age, and education. Assess model fit with a binned residual plot and a calibration plot. Also report the discrimination of your model as summarized by AUC

I fit a model using the following code:
#fit model to predict 10 year probability of heart disease given cigarettes per day, sex, age, and education
model <- glm(as.factor(TenYearCHD) ~ as.factor(currentSmoker) + as.factor(male) + age + as.factor(education), data = Framingham, family = binomial)
summary(model)

and the output was shown in the snapshot

image
image2532×1440 367 KB

Can someone help me understand what summary is asking for and how to run binned residual plot and calibration plot?

Thank you

2

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.