Sei sulla pagina 1di 10

I "'

..
EC05416
Midterm I: Take-Home Component
Fall 2004
All answers must be hand written. No photocopied or printed pages will be accepted.
1.
.---
Let Y denote an nxl vector, Xan nxk matrix of rank k, and pa We observe
Yand X, and wish tn choose Il to minimize the ..sumr' Q(lll=(Y-XIII '(Y-Xlll.
a. Show that
b.
c.
XI y + P XI XfJ)
- 0 t -1 x'y z.x
'
.ll5
-t-
H(fJ) =: ifQ(fJ) -:t JX(l\
iJMP' <:f>
Suffi . nt conditions for to be the global minimizing value of pare: S(P)=O and
A
H ) positive definite for all p. Show that p=(X'XY' X'Y satisfies these
conditions.
The distribution function of a random variable, X, is given by
for A< 0
for 0 A< Y2
for Y2 A< 1
for 1 A
a. Is X a discrete or continuous random variable? Why?
b. Find the density function of X and sketch its graph.
c. Find the expected value of X.
d. Find the probability that X is less than 1/4.
r----=-_-----------------
Assume that a random variable. Z, has MGF, Mz(t). Consider the linear transformation
Y==a+pZ.
a. Express the MGF ofY in terms of the MGF ofZ.
b. Assume that Z-N(O.I). Find the MGF ofZ.
c. Let X==\-L+oZ. l'se your answer to (a) to find the MGF of X.
--
------,==---
j
----. -

--
. .)
I
CAJ (10') (y - Xj'Jy ('y -
=(1) -PiX' )(1- xi
:.y'y_

")( 1,
" I / / '"
- y f> )( I' - I )( I Y r )(",x/1
&d (p)- ') 'y - 2.p'X/ 'I t- (>I )( I Y(!>
-"---
'.
. .
"._.
\.) G.. x.. (, I vJI ve b 00/ (pE'C1-
tb 0- j'j.J:-
) :;/;) :z..X/;<} IJ J(..X)L
..
-- - ----_. --- ---- ----- --- . -_.-
t(t\}
z..-
C) )(l.. <::)
r ( l\J-=. tt x. D 'l\L '/ z
t l/-4x 'h. \ \ --,;
D )
X
I}z. . I
a-e-') E (X') -:. JJ( RNdX .\- 5)< ()l) c1Jl
D l)t
l} 2.. )
::. 5)('1Xd.1I.
:J IJ z.
1/2. 1
-:- ) Lj X\h -l- SqJ( - I.h:"2. J)(
.) )h.. I
:.. ..i 11
3
)' It \- [:1.. -X 2. _
:> v z. .3 J
tjl,
: 1H-f 1fz -*) - (2{it -1
N
)7]
1(i) .- - -h)J
- ;:J. ... l(q
2.1 2 i z i Z -,
..
f ., .
12.-
-
- 2 '1
\
..'
---._---
.. ,;<d) ftXCI'j) :"'-FO'l) :: j+f!.)eY<._- - --. __
. - ------
o
I I
T l.
"2-
- -
- jr.,
L'hr"]
r ( ~ ) ~ C ) ~ ~ ~ '12.. =-;{ ~ "2-
F(V'I):: zLt Y=- Z ( ~ ) : : : it ::: -f
l(-{XZ '/4) =IlgJ
e.:s-r
Iii
ftX'-Y
ti
) ~ 54xc.h
;,
lit-]
::: 3 XL)
t.- ;)
-
.'
,:,5.Q ) rn,/t): J
s
: e e"'-Je..1::f-'-lI(.c)Jc
s
b) N (0 J ;)
v
('1/t)::Je.,
-0,;4
.., .1 z.
-= Jt't.._' -"L"l-J
c "h:Tr e. c
2-
. -IJ"'2
p{z)=-Le... 17-
T2-rr
..
----- --
., sc..J .. -- . _ . _ .
I I '
["Y1)1.l{) j e...f:)C r(z)Jit
s
[
.-
., .
~ )' ' '.
I
I I I C,A.f UC-u... ,
EC05416
Midterm 1: Take-Home Component
FaJl2002
b.
a. 1.
The following questions must be completed and turned in at the beginning of the scheduled "in-
class" exam. All answers must be hand written. No photocopied or printed pages will be .. \ I i,< '. \"(.
accepted. t , \ ( "
Show that the rank of an idempotent matrix equals its trace. 'Hr' k:
o-nd c\ \e 3
Let X and Y denote (nxl) vectors, and Z a scalar. Find aZ/aX' where Z=f(Y) and
Y=g(X) are differentiable functions. \n c.l Co SS
o The distribution function ofa random variable, X, is given by J, .\.,lc \ '0'"
for A < 0
forO A<
for A < 1
for 1 A
/
'/
.;
a. Is X a discrete or continuous random variable? Why?
b. Find the density function of X and sketch its graph.
c. Find the expected value ofX.
d. Find the probability that X is less than 1/4.
The density' function of a Poisson random variable is
fx(x) = (fJ.xei)/xl for fJ.>0 and x = 0, 1, 2, ...
Note that the Poisson is a discrete random variable.. (:\61,") \-he r"
a. Find Fx(A) and show that Fx(oo) =I. ....-t' s'vcl\ghl Q.rvhct.\.hl>"
b. Show that the MGF of the Poisson is Mx{t)=exp(J.L(e'-I)].
c. Find E(X) from the MGF.
Theorem
Given a constant, a, the series
converges to e
a
as n- 00.
ECOS416
1. Define the following terms:
a. random process
b. elementary outcome
c. sample space
d. event
e. the probability ofan event
2. State Kobnogorov's Axioms of Probability.
Homework 1
3. Use Kolmogorov's Axioms and the properties of sets to show that, in general,
P(AB) P(A) P(A+B) P(A)+P(B)
4.
5.
6.
7.
8.
Define the conditional probability of an event A given an event B has occurred. Show c. )
that AIB)+P{ACIB)=:I andP AIB)P(B)+P(AIBC)P(B -P(A). PlAl?)}ltl)r r(t/tJ)trt)
)... :;:.. f<AtH GBl P(IlQ) 4- re/H>
Define the statistical 0 events and B. Sho that statistical p{ JW) r(
independence of A and B implies P(AB)=P(A)P(B). Show that P(AIB)=P(A) implies =t
'
(I: )

A fair die is rolled twice. Assume that the rolls ofthe die are statistically independent.
The elementary outcomes ofthis process consist ofordered pairs. We are interested in
the event AIC:{ "The sum of the two die rolls is five." }.
a. Find the sample space ofthis process. How many distinct outcomes are there?
b. Find the set of ordered pairs that constitute the event A.
c. Let B={"The first roll is one."}. Find the probability that the sumofthe rolls was
five given that we know the fust roll was one. Are the events A and B
statistically independent?
Use the properties ofconditional probability to argue that if fifty-two persons stand in
line and draw, without replacement, a card from an ordinary deck, all individuals have an
equal chance ofdIawing the ace ofspades.
On a recent trip to the Western Economic Association Meetings in Las Vegas, Professor
Zuehlke observes Arnold Zellner playing a slot machine, losing 99 consecutive times,
and stomping offin disgust Professor Zuehlke immediately races from the bar to play
the machine. Based upon years ofexperience teaching introductory econometrics, he has
concluded that he cannot lose since the probability of 100 consecutive losses is virtually
zero. Comments?

Potrebbero piacerti anche