9-2and the values of 0 and 1 are 0 0 and 1 {: }, respectively. The values
of and
2 that maximize Lin 1 are the usual maximum likelihood estimates for and
2 :Substituting these values in L, we haveTo maximize Lin 0 we simply set 0 and then find the value of
2 that maximizes L.
This value is found to bewhich givesThe likelihood ratio isNow sincewe may write as(^) c
1
1 a
1
n
1
bc
1 x
022
s^2 n
d
s
n 2
(^) d
1
1 c
n 1 x
022
g 1 xi x 22
da
n
1
n
1
b
t
n 2
(^) d
1
1
n 1 x
022
g 1 xi x 22
t
n 2
a
n
i 1
1 xi
(^022) a
n
i 1
1 xi x 22 n 1 x
022
L 1 02
L 1 12
c
g 1 xi x 22
g 1 xi
022
d
n 2
L 1 02 c
1
12 n 2 g 1 xi
022
d
n 2
e^1 n^22
ˆ^2
1
n^ a
n
i 1
1 xi
022
L 1 12 c
1
12 n 2 g 1 xi x 22
d
n 2
e^1 n^22
ˆ^2
1
n^ a
n
i 1
1 xi x 22
ˆ
1
n^ a
n
i 1
xi x
PQ220 6234F.CD(09) 5/15/02 8:21 PM Page 2 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH114 FIN L:Quark F