This preview shows pages 1–6. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: The University of Hong Kong
Department of Statistics and Actuarial Science
Stat 1301 Probability & Statistics Assignment 3 suggested solution Section A
$223qu :1 = 0.2 + :11 + 0.3 +f{4}l +02 + 6.1 :2» fig} = 01 b)
x 1 2 3 4 5 6
[51.“ F(3.7)=F(3)=0.6 2.
a) gag +x)"‘2dx = 1,—c(1 +5415: = 1 => c = 1 . .. A" —2~ ‘ v a” 1 __1 ,
b)PCXixIl—fo(1+u3 Gu='C1+u) 0=1~E_E=;~,x:1
c)P£1 «if. x a: 2) = Ea +522ch = ...{1 + ,0 ~21}? =§
3.
8) fo) = PUX] S x) = PlC—x g‘X :1 x) = FCx) ~Fﬂx} b)
F (1‘3 = PW 5’— x) = PlI—v'i a: X 5:. VI?) = FEE} — Fitm 4 a) Find the roots of the polynomial .2: —~ x2 = 0
Possible values for c2 = 0 and ,8 = 1 b) f;c{xx2)dx=1,(xg~§)l% = LC = 5 5. 6. See example class 7 7.
a) Joint cdf F(x,y) b) Marginal density f(x), f(x=‘l)=0.2, f(x=2)=0.5, f(x=3)=0.3 Marginal density f(y), f(y=1)=0.3, f(y=2)=0.26, f(y=3)=0.25, f(y=4)=0.19
f{1,1} =: 0.05 ‘7‘: f(x = :0va = 1) = 0.05 X and Y are not independent. C) X 1 2 3 _
_ F Y) 0.3 0.56 0.81 8. See Example Class 11 1. B1 (a)
(“) (If )
1P(X = ac) = w (“Lb—m ,a; = 0,1, ,Inin{k,a}.
( k )
(b) Note that k and :1: are ﬁxed when a and b diverge to inﬁnity.
IP’(X = m)
a! b!
__ m!(aw)! (Ica:)!(bk+:c)!
'— (a+b)!
k!(a+bk)!
k! (a—I—b—k)! a! b' m!(k—:c)! (a—f—b)! (a—x)!(b—k+ 3:)!
(k)a(a——1)(a,93+1)b(b—l)~(b——k+a:+1)
x (a+b)(a+b—1)(a+b—k+1) k a (1—1 a—x+1
:1: a+b a+b—l a+b~w+1
\._.__._.____.\,._._______/ .._____. a+b—m a+b—a:——1 a+b—k+1
s._______‘,—_—_/ ..) (mm—mm a: ll Comment: The asymptotic distribution of X is a binomial distri—
bution as a and b diverge to inﬁnity.
2. B2 Let N (t),t 2 0 be a Poisson process of rate A > 0, T1 the time of
the ﬁrst event, and for n > 1, Tn denote the elapsed time between the 3. (n — 1)th and nth events. {Tmn = 1, 2,   is called the sequence of
interarrival times. The event {T1 > t} takes place if and only if no event of the Poisson
process occur in the interval [0,t] and thus lP’(T1 > t) = lP’(N(t) = 0) = exp{—At}.
Hence, T1 has an exponential distribution of rate A. Since lP’{T2 > tITl = s} = lP’(O event in(s, 8 +t]lT1 = s) = lP’(O event in(s, 8 + t]T1 = .9) (independent increments) exp{—At} (stationary increments), we have
IP’(T2 > t) = 1E[1P’{T2 > tT1}] = exp{—At}. Therefore, T2 is also an exponential random variable of rate A and, fur
thermore, that T2 is independent of T1. Repeating the same argument
yields Tn, n = 1, 2,    are iid exponential random variables of rate A. B3 (a) It is readily seen fX,y(x, y) 2 O, and +00 +00
/ fX,Y($7 y)dwdy
__ 2 _ 2 2
(a: 2P1!) +(1 My hwy 1:0 “:00 1
: ———————eX ~
/_oo Lo 27n/1 ~p2 pl 2(1p2) +00 2 +00 __ 2 m «276 90 1/27r(1p2) 2(1—p2)
+00 1 y2
= 1. Thus ny is the joint density function of (X, Y). 4. (b) The marginal density function of X is PYCU) = _+me,Y($7y)d$
_ “puma mw m
_ Lo 27n/1—p2 p{ 2(1—p2) }d
_ “Lex 223 1 +°°ex lac—WV m
_ m 13{2} 27r(1~p2)/.oo p{ 2(1~p2)}d
= ﬁexp{%—} 1
_ Thus Y N N(0,1).
Symmetrically we have X N N(0, 1). B4 (a) For any subset A1, .   ,An E R,
IP{X1 6 A1, ,Xn 6 An} 2 Z P(X1=£E1," ,Xn=xn) sue/11 ZnEAn $16A1 wnEAn i=1 931641 “’T‘EA"
i=1 Thus X1,  '  ,Xn are independent.
(b) Not necessarily. Refer to Lecture Note Section 7.2.5, pp.44. Section B 5. Let X be the random variable of the serving time. Let S be the time that the customer at A have
been served when Mrs. Lee arrives at counter B. Probability that counter A will be free after time t P[f;t’c:s—:t}nﬁﬁ>s}:l __' ' = PHI :4; 3 + :1}: > s) = Mij [1 — {Matti} — (1 —~ 8 '35} = 1. — 6'7“ = P(X «a t] =Probability of counter B will be free after time t. Due to the property that the exponential distribution is memory—less, the probability that counter
A will be free is the same as the counter B will be free. 6a)Note that then umber of arrival between [0,t] follows Poisson (At) Let X=# of buses arrival between 10;] Let X1 = number of full bus bemreem It}, 1:], X 2 = number of non.  full bus between. [0, T] (1'er ” 22.! P(X = n} = 9"“ PUQ = m) = PCX = j, m buses are fail} = 2 PM = j} pmﬂw pﬂ’m
Fm Fm PM 233°34er ‘4‘? ji ._m pm m’ (ﬁbril —p3)j _ T y _m
= j: 9 (magwmgujmwy =35; gum): ‘9 A (1"?) J 2m '3 Hm 29mm —— Wife (m1 it?) mi L3 z=c§ :2 P ( ’17) 9.9155 (1,1,{1 _ eQ’ﬁCIQﬁ‘ mi (1 fp mm, — 19)) ___: _______________________ g—h‘p
mi {imp “1
mi glztp Similarly para z m} z 34st,”;
(itr?“
‘ 8
m. b)P(X = X1 =m) = PUL' = m, all are full?) = “2:? p m ...
View
Full
Document
This note was uploaded on 04/29/2010 for the course STAT 1301 taught by Professor Smslee during the Spring '08 term at HKU.
 Spring '08
 SMSLee
 Probability

Click to edit the document details