Method of Moments
Method of Moments
Method of Moments
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Introduction to the Science <strong>of</strong> Statistics<br />
The <strong>Method</strong> <strong>of</strong> <strong>Moments</strong><br />
A (↵, ) random variable has mean ↵/ and variance ↵/ 2 . Because we have two parameters, the method <strong>of</strong><br />
moments methodology requires us to determine the first two moments.<br />
E (↵, ) X 1 = ↵ and E (↵, ) X 2 1 = Var (↵, ) (X 1 )+E (↵, ) [X 1 ] 2 = ↵ 2 + ✓ ↵<br />
◆ 2<br />
=<br />
Thus, for step 1, we find that<br />
↵(1 + ↵)<br />
2<br />
= ↵ 2 + ↵2 2 .<br />
µ 1 = k 1 (↵, )= ↵ , µ 2 = k 2 (↵, )= ↵ 2 + ↵2 2 .<br />
For step 2, we solve for ↵ and<br />
. Note that<br />
µ 2 µ 2 1 = ↵ 2 ,<br />
and<br />
So set<br />
to obtain estimators<br />
µ 1<br />
µ 1 ·<br />
µ 2 µ 2 1<br />
ˆ =<br />
¯X = 1 n<br />
µ 1<br />
µ 2 µ 2 1<br />
= ↵/<br />
↵/ 2 = ,<br />
= ↵ · = ↵, or ↵ = µ2 1<br />
µ 2 µ 2 .<br />
1<br />
nX<br />
X i and X 2 = 1 n<br />
i=1<br />
nX<br />
i=1<br />
X 2 i<br />
¯X<br />
X 2 ( ¯X)<br />
and ˆ↵ = ˆ ¯X (<br />
=<br />
¯X) 2<br />
2 X 2 ( ¯X) . 2<br />
dgamma(x, 0.23, 5.35)<br />
0 2 4 6 8 10 12<br />
0.0 0.2 0.4 0.6 0.8 1.0<br />
x<br />
Figure 13.1: The density <strong>of</strong> a<br />
(0.23, 5.35) random variable.<br />
201