2
Linear Least Squares
The linear model is the main technique in regression problems and the primary
tool for it is least squares fitting. We minimize a sum of squared errors, or
equivalently the sample average of squared errors. That is a natural choice

Problem Set 2
Due in class on paper, Tuesday October 25, 2016
The data
This data set came courtesy of Mark Rober. I mentioned his video
https:/www.youtube.com/watch?v=k-Fp7flAWMA
in a class on experimental design. A student in that class knew him and
intr

Problem Set 4
Due Thursday November 17
Problem 5 is for teams of 1 or 2 or 3. The others are individual effort.
1. Collaborating predictors.
Construct a data set with 20 cases, three predictors X1 , X2 , X3 and a response Y , such that: no regression of Y

\
Stat 305A syllabus
Below is a list of topics for the course. We will get through the first 5 blocks of
topics. We will see some but not all of the last two blocks. Hence the hat on
the syllabus.
This course is about connecting real world problems to sta

Problem Set 1
Due in class on paper, Thursday October 13, 2016
Meta
The class web page has a rubric on how your problem sets should be and
another one on how graders should grade.
R
This problem set includes a gentle introduction to R. If you are new to R