ST3451: Applied Linear Statistical Methods I
Michaelmas Term 2017
Lecturer: Rozenn Dahyot
Timetable
 Monday 11am12 LB107
 Monday 56pm LB107
 Wednesday 23pm LB08
Foreword
5 ECTS module corresponds to about 100125 hours of study time. This includes the 30+ hours of lectures organised on campus that students have to attend. Other sources of information are available to help you (e.g. online ressources such as webpages, videos, MOOCs, and books in TCD libraries). Be creative and independent to build up your skills!
Contacting the lecturer
Emails from students in my taught modules may be ignored for logistic and time management reasons. If so, please ask questions during classes or at the end.
Module content
Simple Linear Regression  Multiple Regression  Regression Diagnostics  Generalised Linear Models
References
There are a lot of books in the Library that are relevant to the course.
Lecturenotes
It is advised to take notes in class. Scans of hand written lecturer 's notes during classes are available on Blackboard. Case studies and R code are also available here.
Weekly timeline

W1: Introduction, Linear Regression
 Dirac Function
 Matrix Calculus
 Chapter 1 of ST3452 Lecturenotes

W2: Multiple Regression : Least Square Estimate, and Standard error
 R code example Multiple Regression with lm() and glm()

W3: Multiple Regression
 AIC short notes Chapter 5
 R code example Multiple Regression using linear algebra
 no class Wednesday 11/10

W4: Exponential family of distributions
 no lecture Monday 16th October due to Ophelia storm
 Distributions for response variable y, read Chapter 2 lecturenotes
 Exercise with distribution
 W5: Case Study: surgery using Bernouilli distribution with logit link function
 W6: (Monday bank Holiday)  tutorial optimisation
 W7: Reading week
 W8: Distributions and Survival Analysis:

W9: Survival Analysis (Further reading Dobson's book chapter 10)
 Case study, 20/11/2017
 Exercises survival analysis, 22/11/2017
 short lecturenote Chapter 8

W10: Deviance
 short lecturenote Chapter 6
 Chisquare Distribution table
 Dirac Distribution (wikipedia)

W11: Poisson Regression  Conclusion
 short lecturenote Chapter 4
 Case Study: Beetles, Binomial and Poisson Regression
 W12: Exam Revisions (see Blackboard): Question time (lecturer available in lecture room for questions from students) Wednesday 13 december 2017: 'question time' postponed to 2018
Exam/Assessment
 Exam (100%)
After exam results
Read What if I want my examination rechecked? and What if I wish to appeal my results?. Request to see script should be made to the SCSS Teaching Unit ( teachingunit @ scss.tcd.ie ) with the following information:
 Exam number of the student
 Code of the module (e.g. ST3010)
 Exam session (e.g. Trinity term, Michaelmas term (supplementals))
Note that marks can not be haggled or changed for other reasons than specified in TCD calendar: a third party (staff member other than lecturer) will also be there monitoring this meeting. Such meeting is not an opportunity for a private tutoring session nor for getting tips for supplementals.