STAT 512
Introduction and Background Review
1
'
$
Class Information
Instructor: Chong Gu
Oce Hours: 1:302:30 MWF, or by appointment, at HAAS 170
Slide 1
Course Work: Homeworks (10%), SAS assignments (10%),
quizzes (10%), midterms (2 20%), and the nal (30%
STAT 512
Simple Linear Regression
'
1
$
Relations Between Variables
Functional relations: Y = f (X ) deterministic, such as (i)
A = r2 for the area A and radius r of a circle; or (ii)
Y = 5 (X 32) for thermometer readings X o F and Y o C .
9
weight (gm)
1
STAT 512
Matrix Approach to SLR
1
'
$
Basic Denitions
Consider a 3 2 matrix
0
1
B
B3
A=@
5
Slide 1
Matrix is an array of numbers.
0
1
a1,1 a1,c
B
C
.C
B.
.C
A=B .
.A
@.
ar,1 ar,c
1
2
C
4C,
A
6
whose transpose is given by
0
1
A = @
2
3
5
4
6
1
A.
Vector is
STAT 512
'
Slide 1
Analysis of Covariance
One possible approach is to take
the weight dierence and do a 1way ANOVA. An alternative approach is to use ANCOVA with
Y the nal weight and X the initial weight.
&
calculate
SSR(A|X )/(a 1)
,
SSE(A, X )/(n a 1)
a