Engineering Probability Class 15 Thurs 2021-03-18
Table of contents::
1 No class next Thurs 3/25
Attend Pres Jackson's spring town meeting, or sleep late, or go on a hike, or whatever you want.
Homework that was due then is postponed.
2 Chapter 4, ctd
Example 4.22, page 169.
-
Section 4.4.4 Gamma r.v.
Setting its 2 parameters to various values gives useful special cases.
E.g. m-Erlang, the sum of m exponential r.v.
-
Section 4.4.5, Beta r.v., p 173.
Setting its 2 parameters to various values gives useful special cases.
-
Section 4.4.6 Cauchy r.v.
has no moments.
Problem 4.17, p 217.
Problem 4.27, p 218.
Problem 4.39, p 219.
Problem 4.56a, p 220.
Problem 4.68, p 222.
3 Chapter 5, Two Random Variables
-
One experiment might produce two r.v. E.g.,
Shoot an arrow; it lands at (x,y).
Toss two dice.
Measure the height and weight of people.
Measure the voltage of a signal at several times.
The definitions for pmf, pdf and cdf are reasonable extensions of one r.v.
The math is messier.
The two r.v. may be *dependent* and *correlated*.
The *correlation coefficient*, $\rho$, is a dimensionless measure of linear dependence. $-1\le\rho\le1$.
$\rho$ may be 0 when the variables have a nonlinear dependent relation.
Integrating (or summing) out one variable gives a marginal distribution.
-
We'll do some simple examples:
Toss two 4-sided dice.
Toss two 4-sided ''loaded'' dice. The marginal pmfs are uniform.
Pick a point uniformly in a square.
Pick a point uniformly in a triangle. x and y are now dependent.
-
The big example is a 2 variable normal distribution.
The pdf is messier.
It looks elliptical unless $\rho$=0.
I finished the class with a high level overview of Chapter 5, w/o any math.