COURSE OUTLINE

STAT 151A: LINEAR MODELLING - THEORY AND APPLICATIONS

FALL 2015
UNIVERSITY OF CALIFORNIA, BERKELEY

Instructor: Aditya Guntuboyina. Email: aditya@stat.berkeley.edu and Website: www.stat.berkeley.edu/~aditya

Lectures: 11 am to 12:30 pm on Tuesdays and Thursdays at 2 LeConte Hall.
Office Hours: 9:30 am to 10:30 am on Tuesdays and Thursdays at 423 Evans
Hall.
GSI: TBA. Email: TBA
GSI Lab Section: 12-2 pm or 2-4 pm at 340 Evans Hall on Wednesdays.
GSI Office Hours: TBA
Short Description: Linear modelling is the most important tool to solve regression
problems where the focus is on understanding the relatioship between a response variable
and a bunch of explanatory variables. These problems arise in a very diverse set of
applications. In addition to being practically invaluable, linear models also have a rich
and beautiful mathematical theory. The main goal of this class is to understand this
theory and to learn how to do applied regression analysis via linear models. I will also
spend some time on linear methods for dimension reduction such as principal component
analysis.
Tentative List of Topics: The linear regression model, Estimation, Inference, Diagnostics, Applications to ANOVA models, Variable Selection, Generalized Linear Models,
Principal Component Analysis.
Prerequisite: Atleast one previous statistics class (ideally STAT 135) is necessary. This
class will be heavy on linear algebra. I will review the basic concepts from linear algebra
when they will be used but prior experience is necessary (ideally you would have taken
MATH 110 but MATH 54 is also okay if you work hard). Also required are basic calculus
(MATH 53) and elementary probability (STAT 134).
Text: There is no required textbook. But the following books are good and you may use
them as references. There are also many other books and freely available material online
for the materials in this course.
(1) Linear models with R by Julian Faraway
(2) Extending the linear model with R by Julian Faraway (for the second part of the
course on generalized linear models)
(3) Plane Answers to Complex Questions by Ronald Christensen. This book, which
is available for free via the library website, covers the theory aspects very well;
but it is a little dry.
1

COURSE OUTLINE STAT 151A: LINEAR MODELLING - THEORY AND APPLICATIONS FALL 20152

(4) Generalized Linear Models by McCullagh and Nelder (for the second part of the
course). This is a slightly advanced book however.
(5) An Introduction to Statistical Learning by James, Witten, Hastie and Tibshirani.
This is one of the texts used for STAT 154. Chapters 3, 4, 5, 6 and 10 are relevant
to this course. This book is available for free via the library website.
(6) Introductory Econometrics by Wooldridge. The first part of the book provides an
excellent treatment of linear regression analysis.
(7) Statistical Models by Freedman. I think that this is a great book to read once you
know the mathematical basics of linear models. The preface states this lively and
engaging textbook explains the things you have to know in order to read empirical
papers in the social and health sciences, as well as the techniques you need to build
statistical models of your own.
Lecture Notes and R code: I usually prepare typed notes for each lecture. I also show
data analysis examples on R in class. I will post both the lecture notes and R code on
bcourses after each lecture.
Homework assignments: These will involve a mix of theoretical (pen and paper) and
computer exercises. I will post them on the class bcourses page (roughly once every two
weeks) on a Tuesday and they will be due at the beginning of the Wednesday section
eight days later. Late assignments will not be accepted.
Exams: There will be three exams in all: two midterms and one Final. Midterm 1 will
be an in-class exam on October 8 (tentatively) and Midterm 2 will be either an in-class
exam or a take home exam (if in-class, it will be on November 17 tentatively). The Final
is on 16 December from 8 am to 11 am.
An important policy about the Final: Please drop this class if you are taking another
class whose final exam is also from 8 am to 11 am on 16 December.
Assessment: Your final grade will be a weighted average of homework (30%), Midterms
(35% = 20% of the midterm you do well in + 15 % of the other midterm), final (30%)
and piazza participation (5%).
Grade Complaints: If you have a complaint against an assigned homework or exam
grade and want to talk to me about it, first send me a written request through email
explaining your case clearly.
Academic Integrity: You are encouraged to work in small groups on homework problems. However, you must write up solutions on your own, and you must never read or
copy the solutions of other students. Similarly, you may use books or online resources
to help solve homework problems, but you must credit all such sources in your writeup
and you must never copy materials verbatim. Any students found to be cheating automatically risks failing the class and being referred to the Office of Student Conduct. In
particular, copying solutions, in whole or in part, from other students in the class or any
other source without acknowledgement constitutes cheating.
Students with disabilities: If you need accommodations for any physical, psychological,
or learning disability, please get in touch with me so that we can make the necessary
arrangements.