Scribd
Upload
14 views1 page

10 A 5

10a5

Uploaded by

PETER
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views1 page

10 A 5

10a5

Uploaded by

PETER
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Department of Statistics

COURSE STATS 330/762

Assignment 5, 2010

Instructions: Hand in your completed assignment to the Student Resource Centre by 4pm
October 14th.

The data set for this assignment is in the file acath.txt which is available on the course
web page.

These data are from a medical study of individuals who had exhibited symptoms of
coronary artery disease. The response variable tvdlm is an indicator of whether or not
that patent, when examined, actually has severe coronary artery disease (Three-vessel
or left main disease). It has value 1 if severe disease is found, and 0 otherwise.

The other variables in the data set are

sex: gender , scored as 0=male, 1=female


age: Age in years
cad.dur: Duration of the symptoms (weeks)
choleste: Blood cholesterol reading

1. Read the data into R, and make a data frame. Check for gross errors. (Note
that, due to the size of this data set, I have not reproduced it at the end of
the assignment. I have not deliberately introduced any errors into the data. )
Print out the first 16 lines. [5 marks]

2. Perform a graphical analysis of the data, without fitting any model, that will
let you see how the risk factors sex, age, cad.dur and choleste affect the
probability of having severe coronary artery disease . [10 marks]

3. Fit a logistic regression model to the data, diagnosing any major problems
with the fit. Interpret the coefficients. Does your interpretation confirm the
conclusions you reached in question 2? [15 marks]

4. Produce a graph that will illustrate how the fitted probabilities for your model
change with age and sex. (You can fix the values of cad.dur and choleste at
their averages.) Is the risk higher for males? [10 marks]

5. Extra question for 762 only: Do you think the model you have chosen has any
value as a predictive tool?

You might also like