Biostatistical Methods II:  Regression Analysis (Biometry 701)

Spring 2010

 

Description:  This is a one-semester course intended for graduate students pursuing degrees in biostatistics and related fields such as epidemiology and bioinformatics.  Topics covered will include linear, logistic, poisson, and Cox regression.  Estimation, interpretation, and diagnostic approaches will be discussed.  Software instruction will be provided in class in R.  Students will be evaluated via homeworks (55%), two exams (35%) and class participation (10%).  This is a four credit course.

 

Textbooks:

(1) Introduction to Linear Regression Analysis (4th Edition).  Montgomery, Peck and Vining.    Wiley; New York, 2006.

(2) Regression with Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis.  Frank E. Harrell, Jr.  Springer; New York, 2001.

 

Prerequisites:  Biometry 700

 

Homeworks Policy:   Homeworks are due by 5pm on the due date.  Asking for extensions on homeworks is strongly discouraged.  It is not fair to other students.  However, it is expected that on occasion extenuating circumstances may arise.  Therefore, the policy is that each student may request a extension on homework twice and the extension is to be no more than 2 days.   After using two extensions, no more extensions will be granted except with a medical note.  

 

Course Objectives:  Upon successful completion of the course, the student will be able to

    1.  Apply, interpret and diagnose linear regression models

    2.  Apply, interpret and diagnose logistic, poisson and Cox regresssion models 

 

Instructor:

Elizabeth Garrett-Mayer

Website:

http://people.musc.edu/~elg26/teaching/methods2.2010/methods2.2010.htm

Contact Info:

 Hollings Cancer Center, Rm 118G

 

garrettm@musc.edu (preferred mode of contact is email)

 

792-7764

Time:

Mondays and Wednesdays, 1:30-3:30

Location:

Cannon 301, Room 305V

Office Hours:

Tuesdays 2:00-3:30, or by appointment

 

Lectures:

Date

Topic

Lecture Notes

Computing

Readings (Texts referred to as MPV and Harrell)

 

 

 

 

 

W Jan 6

Introduction to regression; simple linear regression (SLR)

lect1.ppt; ls.estimates.pdf

R code

MPV 1, 2.1-2.2; Harrell 1

M Jan 11

Linear model properties; software tutorial

lect2.ppt

 

Harrell 6; R-intro.pdf (supplemental)

W Jan 13

Inferences and assumption checking in SLR

lect3.ppt;

R code

MPV 2.2, 2.3, 2.5; Harrell 9

M Jan 18

no class!  MLK Jr. Day

 

 

 

W Jan 20

SLR Diagnostics

(see Jan 13 notes);

R libraries.ppt

 

MPV 2.4, 2.5, 4.1, 4.2

M Jan 25

Inferences in SLR

lect4.ppt

R code

 

W Jan 27

Diagnostics (continued); Correlation; Intro to Multiple Linear Regression

lect5.ppt

R code

MPV 3.1, 4.4, 5

M Feb 1

Hypothesis testing in MLR

lect6.ppt

R code

MPV 3.2-3.5, 3.7

W Feb 3

MLR

lect7.ppt

R code

MPV  4.2.4; Harrell 2.1-2.2

M Feb 8

Forms for predictors

lect8.ppt

R code

MPV 7, 8; Harrell 2.3-2.4

W Feb 10

interactions

lect9.ppt

R code

Harrell 2.3

M Feb 15

F-tests and ANOVA; F-tests and Coefficients of Determination

lect10.ppt

R code

MPV 2.3.3, 2.6, 3.3, 4.5; Harrell 2.7

W Feb 17

Multicollinearity

lect11.ppt

R code

MPV 3.9,11

M Feb 22

MLR Model Building

lect12.ppt

R code

MPV 3.9, 9, 10;  4.2.4; Harrell 4

W Feb 24

MLR diagnostics

lect13.ppt

R code

MPV 6

M Mar 1

MLR:  example (Harrell Ch 7)

(see lect13.ppt)

 

Harrell 7

W Mar 3

Exam 1

 

 

 

M Mar 8

NO CLASSES:  SPRING BREAK!

 

 

 

W Mar 10

NO CLASSES:  SPRING BREAK!

 

 

 

M Mar 15

Introduction to logistic regression; link functions

lect14.ppt

R code

MPV 14.1-14.2; Harrell 10.1

W Mar 17

Likelihood ratio tests and deviance

lect15.ppt

R code

Harrell 10.2-10.4

M Mar 22

Goodness of fit, Information criteria, ROC analysis

lect16.ppt

R code

Harrell 10.3-10.9

W Mar 24

 

 

 

 

M Mar 29

Logistic regression for case control studies

lect17.ppt

 

 

W Mar 31

Logistic regression example

(see lect17.ppt)

 

Harrell 11 or 12

M Apr 5

Ordinal logistic regression

lect18.ppt

stata code

R code

Harrell 13.1-13.3

W Apr 7

Introduction to survival data

lect19.ppt

 

Harrell 16.1, 16.2

http://jslhr.asha.org/cgi/reprint/42/2/432

http://www.walkerbioscience.com/pdfs/Survival%20analysis.pdf

M Apr 12

Survival analysis

 

 

Harrell 16.5

W Apr 14

Cox regression

lect20.ppt

R code

Harrell 19

M Apr 19

Cox regression examples

aje paper

jco paper

 

Harrell 20

W Apr 21

Poisson regression

lect21.ppt

R code

 

M Apr 26

Random effects models

lect22.ppt

 

 

 

Datasets:

SENIC data:  senicfull.csv; codebook

Ischemic Heart Disease:  ischemicheartdisease.csv; codebook

ICU data:  icu.csv; codebook

YTS:  yts.subset.2007.csv

kidney dialysis:  kidneydialysis.csv

 

Homeworks:

Homework 1:  due 01/20/10

Homework 2:  due 02/03/10; key for MPV 2.2.1

Homework 3:  due 02/15/10

Homework 4:  due 3/31/10

Homework 5:  due 4/19/10

Homework 6:  due 5/5/10

 

EXAMS:

 

Articles:

Patel et al. NEJM, 362(10)

hanley&mcneill.pdf:  Comparing AUCs for ROC curves based on the same data

gillison.pdf:  Case control study of HPV16 and Oropharyngeal Cancer

chaves.pdf:  Nomograms for mobility disability

berenholtz.etal.pdf:  Poisson regression example

 

 

 

Computing: 

    R website:  http://cran.r-project.org/

    R tutorial:  R-intro.pdf

    Stata website:  http://www.stata.com/