Bootstrap
Lecture 20
Dr. Mine Çetinkaya-Rundel
Duke University
STA 199 - Fall 2022
Invalid Date
Announcements
– HW 5 (Due Thursday)
– Project Proposal (Due Friday)
Project Hints
– Reference the AEs and Videos for example research questions
– Make sure your variable is not an id
– Write yourself a project outline
Goals
– What is bootstrapping
– What is a confidence interval
Warm Up: Remember the Penguins
![]()
\(\widehat{body.mass} = -5781 + 49.7 \times flipper.length\)
– What does this line tell us?
\(body.mass = \beta_o + \beta_1 \times flipper.length\)
What is \(\beta_1\)?
Does our estimated line tell us what \(\beta_1\) is?
Population Parameters
– \(\beta_1\)
– \(\mu\)
– \(\pi\) or p
. . .
This week
– Is all about concepts
– Main goal is to develop a fundamental understanding about bootstrapping, confidence intervals, and hypothesis testing.
We will put our understanding into more practice starting next week.
In Summary
Bootstrap methods are simulation methods
Variability - Uncertainty: How spread out your data are
We can use this to guess plausible values around our parameter of interest
This is called a confidence interval, and it’s different than probability
