Latest news
Lecture Tue 13.15-15 in MVF32 and Thur 13.15-15 in MVH11.
Office hours Fri 13.15-15.00 in my office
Thursday lecture = review lecture, and we'll go over the old final.
Friday lecture = in-class presentations. Please prepare 4-6 slides (just print them out on paper. State the problem and describe the data. Show figures and preliminary results. Any complications/challenges? Analysis plan - be specific to get the best feedback.
Final is Dec 16. Project due date after the break - specific date TBA.

Examiner and lecturer
  Rebecka Jörnsten (jornsten@chalmers.se)

Course literature
Main text book
Draper and Smith, Applied Regression Analysis

Other texts
Lecture notes

Syllabus
(pdf file)
 
 
Preliminary Course Plan

Lectures
Week
Chapter
Contents
  43   0, 1, 2:1-6   Introduction. Simple linear regression, diagnostics.
  44   4, 5:1-4, 6:1-2, 8:1-3   Multiple regression, diagnostics and testing.
  45   14, 23, 9:1 + notes   Dummy variables, ANCOVA, Model selection.
  46   11, 15 + notes   Model Selection criteria
 47   26 + notes   Bootstrap. Cross-validation.
 48   16:1-4, 17   Regularized regression
 49   9:2-3, 9:5, 18  Weighted Least Squares, Non-linear models, Generalized linear models. In-class presentations
 50 - Dec 16     Final Exam

Lecture Notes
Week
   Notes and handouts
 43   Lecture 1 , Lecture 2 , Demo 1
 44   Lecture 3 , Lecture 4
 45   Lecture 5 , ldldemo.r Run demo - read the comments in the file, ldldata The data file

Lecture 6a , Lecture 6b.

Stepwise backward model selection to try on the LDL data: add this line of code to the ldldemo.r file: print(step(mm1b)). This eliminates one variable at a time until a criterion AIC is minimized, then it stops. We will get to AIC next lecture, but for now think about it as something similar to doing a backward F-test.

 46   Lecture 7 - recap , Lecture 7 - model selection.

Lecture 8 - model selection , My R code for lab 2

 47   Lecture 9a - indicator variables, Lecture 9b - crossvalidation

The full pollution data , Exploring the full pollution data, Model selection code, The code that runs model selection several times.

 48   Lecture 11 - CART , CART demo on cars data , cars data for CART demo

Final exam from last year

Lecture 12 - logistic regression , demo code , the wine data

 49   Lecture 13 - regularized regression , demo code , the cars data

Labs and Exercises
Week
   Labs and data links
 43   Lab 1 R-tutorial. No report due.

The animal data

 44   Lab 2 Least Squares part 1. Report due Nov 16.

The pollution data

The potato data

 45  
 46  
 47   Lab 3 Least Squares part 2. Report due Dec 7.

A bit of code to get started , The prostate cancer training data

The prostate cancer test data

 48  
 49  
Examination
To pass this course you must hand in satisfactory lab reports, complete and present a data analysis project, and pass a final exam (details - see Syllabus ).

The exam takes place on Dec 16. Room: To be announced.
The exam is open-book and open notes.
Bring ID and receipt for your student union fee

You will be notified the result of your exam by email from LADOK (This is done automatically as soon as the exams have been marked an the results are registered..)
The exams will then be kept at the students' office in the Mathematical Sciences building.
Check that the number of points and your grade given on the exam and registered in LADOK coincide.
Complaints of the marking should be written and handed in at the office. There is a form you can use, ask the person in the office.).

The following link will tell you all about the examination room rules at Chalmers: Examination room instructions

Computer labs
Please see the Syllabus for instructions on how to write a lab report.