Statistics 102H: MLE's (Statistical Inference by S. D. Lilvey
Statistics 102H: MLE's (Statistical Inference by S. D. Lilvey
Administrivia
pass out evaluations
I emailed comments. If you need your data analysis back, come and get it.
I'll pick out some questions that look like exam questions and put them on the homework page.
Homework
Page 84 of handout:
4.1 just compute the MLEs for each case
4.2
4.3 (skip the "not idenifiable" part)
4.6 (skip simple approximation)
BONUS: 4.7 is hard! It requires doing some geometry.
Review for exam
Science
What does it take to prove a scientific claim?
What are the biases that you need to look for?
Understand the box model: be able to answer if a scientific experiment is a box model or not.
Primary goal: be an intelegent consumer of modern science. Read a story and identify why it isn't real science.
Study: Death certificates of taller people have younger ages. Do people shrink as they get older? Does being talk cause death? Or is there a lurking variable?
Study: Primary end point is AIDS. Vacine doesn't work. But, for a subset (black and latinos) it does have a p-value less than .05. Is it real or not?
Mathematical statistics
Be able to relate probabilistic models to statistics
Identify the parameter
Be able to compute the MLE
Know what the Fisher Information is (i.e. the negative of the 2nd dervative)