PDF of Lecture Notes - School of Mathematical Sciences
PDF of Lecture Notes - School of Mathematical Sciences
PDF of Lecture Notes - School of Mathematical Sciences
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
1. DISTRIBUTION THEORY<br />
where φ(a) = Φ ′ (a) = 1 √<br />
2π<br />
e −a2 /2 is the N(0, 1) <strong>PDF</strong><br />
= 1 2 x−1/2 [ 1<br />
√<br />
2π<br />
e −x/2 + 1 √<br />
2π<br />
e −x/2 ]<br />
(<br />
= (1/2)1/2<br />
1<br />
√ x 1/2−1 e −1/2x, which is the Gamma<br />
π 2 , 1 2)<br />
<strong>PDF</strong>.<br />
On the other hand, the distribution <strong>of</strong> Z 2 is also called the χ 2 1, distribution. We<br />
have proved that it is the same as the Gamma ( 1<br />
2 , 1 2)<br />
distribution.<br />
3. Moments <strong>of</strong> transformed RVS:<br />
If U ∼ U(0, 1) and Y = − log U<br />
λ<br />
then Y ∼ Exp(λ) ⇒ f(y) = λe −λy , y > 0.<br />
Can check<br />
E(Y ) =<br />
∫ ∞<br />
0<br />
λye −λy dy<br />
= 1 λ .<br />
4. Based on Theorem 1.6.1:<br />
If U ∼ U(0, 1) and Y = − log U , then according to theorem 1.6.1,<br />
λ<br />
E(Y ) =<br />
∫ 1<br />
0<br />
= − 1 λ<br />
= − 1 λ<br />
− log u<br />
(1) du<br />
λ<br />
∫ 1<br />
0<br />
log u du = − 1 λ (u log u − u) ∣ ∣∣∣<br />
1<br />
[<br />
u log u ∣ ∣1<br />
0 − u ∣ ]<br />
1 0<br />
0<br />
= − 1 [0 − 1]<br />
λ<br />
= 1 , as required.<br />
λ<br />
There are some important consequences <strong>of</strong> Theorem 1.6.1:<br />
20