28.01.2015 Views

PDF of Lecture Notes - School of Mathematical Sciences

PDF of Lecture Notes - School of Mathematical Sciences

PDF of Lecture Notes - School of Mathematical Sciences

SHOW MORE
SHOW LESS

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 />

⎧∑<br />

h(x)p(x)<br />

⎪⎨ x<br />

E{h(X)} =<br />

⎪⎩<br />

∫ ∞<br />

−∞<br />

h(x)f(x) dx<br />

if ∑ |h(x)|p(x) < ∞ (X discrete)<br />

x<br />

if<br />

∫ ∞<br />

−∞<br />

|h(x)|f(x) dx < ∞ (X continuous)<br />

The moment generating function <strong>of</strong> a RV X is defined to be:<br />

M X (t) = E[e tX ] =<br />

↓<br />

moment<br />

generating fn<br />

<strong>of</strong> RV X.<br />

⎧∑<br />

e tx p(x)<br />

x ⎪⎨<br />

⎪⎩<br />

∫ ∞<br />

−∞<br />

e tx f(x) dx<br />

X discrete<br />

X continuous<br />

M X (0) = 1 always; the mgf may or may not be defined for other values <strong>of</strong> t.<br />

If M X (t) defined for all t in some open interval containing 0, then:<br />

1. Moments <strong>of</strong> all orders exist;<br />

2. E[X r ] = M (r)<br />

X<br />

(0) (rth order derivative) ;<br />

3. M X (t) uniquely determines the distribution <strong>of</strong> X:<br />

M ′ (0) = E(X)<br />

M ′′ (0) = E(X 2 ),<br />

and so on.<br />

1.1 Discrete distributions<br />

1.1.1 Bernoulli distribution<br />

Parameter: 0 ≤ p ≤ 1<br />

Possible values: {0, 1}<br />

Prob. function:<br />

⎧<br />

⎪⎨ p, x = 1<br />

p(x) =<br />

⎪⎩<br />

1 − p, x = 0<br />

2<br />

= p x (1 − p) 1−x

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!