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 DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Chapter 2 Multivariate Distributions 2.1.2 2.1.5 2.1.6 2.1.7 2 $2+$2 7r x2+$2] dIBdm
fa [WI—b/Fj 1 2 II II Why is —10g 2 > 0? H [0: [OW/2i29(p)/7rp]pd9dp f0 9(p)dp= 1. P(X+Ysz)=/ / e_$_ydydx
0 0 z
/  e‘(23)]€—$ dx :21_ 6—2 _ 262.
0 2e” 0<z<oo G’(z) ={ O elsewhere 1 1
P(XY S 2) = 1 “ / / dydx
2 2/3 1/:(1_; 0(2) = { 11 Z d$=Zzlogz
) “iogz 0 elsewhere. 12 Multivariate Distributions 2.1.8 13)(13 ( 26 )
m 13—:—— x Z (Ly 2 0,1 + y g 13,31: and y integers f($ay) = 13
0 elsewhere.
2.1.10
1/2 1—121
P(X1+X2 Sl)=15/ z? 1E2d1122:ld.’131.
0 231
2.1.14
00 oo 1 i+j
E[et1X1+t2X2] : ZZetii+t2j <_>
i=1j=1 2
00 i 00 j
1 1
= z eh) 2e)
1:]. < 2 j==1 2
1 1
= [1 —— 2‘16‘1 — 1] [1 — 2‘16‘2 *1] ’ provided t, < log 2, i = 1,2. 2.2.1 2 2312
p(y1,y2)={ (gay (21:) :leslégil;(0’0)’(*1illi(1a1),“),2) 2.2.2 __ 3/1/36 3/1 = y2,2y2,3y2;y2 =1,2,3
p(yl’y2)  { 0 elsewhere. an
m» was 2.2.4 The inverse transformation is given by $1 = ylyz and x2 = 3/2 with Jacobian
J = y2. By noting what the ‘boundaries of the space 8(X1, X2) map into, it
follows that the space T(Y1, Y2): {(311,312) :0 < y. < 1,2" = 1,2}. The pdf of (Y1, is fY1,Y2(3/1s 3/2) = 2.2.5 The inverse transformation is 11:1 = y1 — y2 and 2:2 = y; with Jacobian J = 1.
The space of (Y1,Y2) is T = {(y1,y2) : —00 < y. < 00,71 = 1,2}. Thus the
joint pdf of (Y1, Y2) is hag/2011,30) = fX;,X2(yl — 212,3!2), which leads to formula (2.2.1). 2.3.2 2.3.3 f2($2)
f12(351l932)
E(X1$2) (3(y) 9(y)
E(Y)
Vhr(ij
13LX1) Var(X1) 22
01/ 231/1173 (11121 =
O 01
2 10m1m§,0 < m; < 932 < 1;zero elsewhere 1/2
f...
1/2 1/4 II II II 64 12 2:55 (21%) 2m1/‘(5/8)2 dm 3 1 1/2 / 10321323 (1:32de 2/ —x1(1— dxl 2
1‘1 1 /4 1132
f0 21x§x3dm1 = 7:33, 0 < 132 < 1. = 55 Zlmfmg/Z'Bg = 3zf/m‘3, 0 < :01 < :32. 3
—.’132. $2
/ m1(32§/mg)dm1 =
0 4 4y/3
P<§ngy)=/ 7$3d$2=< y
0 7QY¢ 0<y<§
0 elsewhere.
7 3 __ 21 §Z‘§'
7
1024'
g1
32'
553 7 “>— 15360 1024 ' 2.3.8 The marginal pdf of X is 00
fx($) = 2/ (fie—y dy = 26“”, 0 < :1: < 00.
Hence, the conditional pdf of Y given X = a: is
26"”6'1’ _ _
fylxlylx) = 264$ = 6 (y x), with conditional mean W
E(YlX = 9:) =/ ye_(y“”) dy = a: + 1, a: > 0. 4 3 _—_1=>c1=23ndc2=5. 0<x<y<oo, 13 ,W
15% QUIZ 1 SOLUTION (VEKSmN M
Di: 7%, COMWWLA (Jam/{wk X Nara 700% j”) = x/é J 7cm Z<><<L{,
H = ijewa z ska/Ma = [79?] = é? d; = EOﬂ ‘ﬂ; = 10” (“$32 =, "E? but Y: VX v 77% dish‘buﬁom mam 01C Y is EM = P(Yég\ = PCXW/g‘) = 5; {wk
T95 some TO quz 2 g 5%) X>0) (3)0}, ><+g<4
’ 4/2 “X V PLXW) r. 3&6 {(xl3>d%>dx 72* ' .1 ,
: Si(g:wq><pa\d><)a(g + 18/2( O ‘*9( “RCA 2 _g: “1 SQ éxdé :: .. 00 $093} w Ex (9%
(820% icio‘) 6% g) ‘4 éxU—x) 0%
ix 1 :32 (_X “X
z 8 % mlﬂx 0W3 ’ — —f104x<1 SOLUTION TO QUIZ 2 £993) '= 60ng)2(><>J Vol x+g< 1 PORVB = $1 {(xlﬂmé) 61X X 11.7. ' = 5 (quxvzﬁdxwg + [g/Xigjw‘wkwj O ...
View
Full Document
 Spring '08
 AlexeiStepanov
 Statistics, Probability, Trigraph, Conditional expectation, Multivariate Distributions, Jacobian, inverse transformation

Click to edit the document details