Stat 200b. Chapter 8. Linear regression models.

1
Stat 200b. Chapter 8. Linear regression models.
2
n by 1, n by 2, 2 by 1, n by 1
3
(No Transcript)
4
13 by 5
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
Effect of increasing seat height is 2?1
9
Some matrix review transpose
multiplication inverse derivatives
10
Normal linear model. Estimation
11
Profile log likelihood
12
Straight-line/simple regression.
13
(No Transcript)
14
Fitted values.
NB. Assuming matrix inverse exists
15
Weighted least squares.
inverse existing
16
Example 8.8. Cycling data
17
(No Transcript)
18
(No Transcript)
19
Example 8.10. Maize data.
20
Likelihood quantities.
21
Take expected values
22
(No Transcript)
23
Normal distribution theory. Full rank case
24
A useful decomposition.
25
Confidence interval.
26
Gauss-Markov Theorem. page 374
27
There is a generalized inverse variant Example.
28
Eg. Teaching methods data, p. 427 Method
average Usual 17
14 24 20 24 24 19.67 Praised 28 30 29 24 27
23 27.41