Professional Documents
Culture Documents
DR. T. R. FIELD
December, 2011
1. There is a collection of field goal kickers, which can be divided into two groups 1
and 2. Group i has 3i kickers. On any kick, a kicker from group i will kick a field
goal with probability 1 /(i 1) , independent of the outcome of any other kicks by that
kicker or any other kicker.
(a) A kicker is selected at random from among all the kickers and attempts one field
goal. Let K be the event that a field goal is kicked. Find P[K ] .
(b) Two kickers are selected at random. For j 1,2 , let K j be the event that kicker j
kicks a field goal. Find P[ K1 K 2 ] . Are K 1 and K 2 independent events? Justify
your answer intuitively.
2. The moments of a continuous non-negative random variable X with density
function f X (x) , x 0 are defined as E[ X n ] for integer n 1 . Construct an example
for each of the following cases:
(a) f X (x) does not tend to zero as x ,
(b) f X (x) as x 0 and all moments exist,
(c) f X ( x) 0 as x and no moments exist.
3. (a) Describe a procedure for generating independent identically distributed (i.i.d.)
samples of a random variable via uniform sampling and the (inverse of the) CDF.
(b) Given a large collection of i.i.d. samples of a random variable, explain a sorting
procedure that would yield (a discrete approximation to) the underlying CDF.
Explain qualitatively the relationship between these two procedures.
Continued on Page 2
Page 2
4. (a) Given a function g ( X , Y ) of a pair of random variables X , Y explain a
general procedure for obtaining the CDF of W g ( X , Y ) from the joint density
function f XY ( x, y) .
(b) Find the CDF and PDF of W XY when random variables X and Y have
joint PDF
0 x 1, 0 y 1
otherwise.
1
f X ,Y ( x, y )
0
t 0,
otherwise.
and compute its moment generating function (MGF). Batches of data are
independently corrupted with probability q . Show that the distribution of the
random variable K equal to the index of the first un-corrupted batch is geometric
with PMF
(1 q )q k 1
PK ( k )
0
k 1, 2, ... ,
otherwise.
and compute its MGF. Consider the random variable V equal to the total waiting time
till arrival of the first un-corrupted batch of data identify its PDF. [Hint: you may
assume that, for X i a collection of i.i.d. random variables with MGF X (s ) and N a
non-negative integer-valued random variable that is independent of X i , the random
sum R X 1 ... X N has MGF R ( s ) N (ln X ( s )) .]
6. (a) True or false: For a continuous-valued random process X (t ) , the random
variable X (t 0 ) is always a continuous random variable. Explain your answer
carefully.
(b) Consider the random process
W (t ) X cos(2f 0 t ) Y sin(2f 0 t )
where X and Y are uncorrelated random variables, each with expected value 0 and
variance 2 . Find the autocorrelation RW (t , ) . Is W (t ) wide sense stationary?
The End