0 ratings0% found this document useful (0 votes) 64 views8 pagesLecture Notes - Week 10
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here.
Available Formats
Download as PDF or read online on Scribd
Oar Contos Otba'bohbns
Werlbll Diski'bubron .
het X be conk. rv Via &>0,fSO ¥ 14 dass
1's en ee
§ “4 one Hp xP te if x0
Kb sed bo have Werlbll did. with param ehers
oe, B- Thee ts a pearral form af Werblt density
uncbion hut in most Phy sleet ay perc edions this
for bused | xk Sas
(mane den 68 Df nd Tt a of|
Leb Xa Weibel (t,p) then tt can be shown that
ca |
EQ." C(LtO and
ee 2
VartX)=2 oF fr(gen-( Gr] f
Alok Bak whee pol for se" > KaenpOne)
Example -
Life bine af a componsat Gb ptediled hu a Whbutl
dist with prt. shale with a (nipe crember of
components , ts observed amon She com nents
flashed aflr 30 hovers, 5Yo of'tham failed before
00 heurs, Find Fhe velve o [remeber &
Xi bibbine Xap Werbl (d, peo) w= ©
(22)2
=o
{exe ZkXE As@
[CX £100 |X>30) = O18
=> PCX 4100 X>90) a 2 (990 2K4/00)
P(X>90) /(X590)
joo i : 7
= aX —< 790) — 0/00)
of 2k£KE "oly ae in ae
oi . ~ Lh Bayz aaa
f, 20K E*ele ge
0
a is _ 090) -<¢100)*
SEXO isjeE = E
—k(90)* kl, 2
=> 7-9/5 )é Di | pent 2)
— A090)* _a00)*
=> (Corse | mu)
=> €n(0,85) oh (90)*= _ of (/00)
= Aha (0,88) = (8/00 — 0090 ) of
Lp 48S)
eth i
=) += ~Xyee0
(60)Zxercise ©
A4-X be a tehme of » component. Far ture
pale of a component 1s $b) iG » whee.
|) & forchor valve of te chasihy of X at
hime, t and RU)= PCXKS€) 1s rehnbhts by
af component at ne t. show that yy
a) Xs exponendynl, furhire rake 1k conchant
LB) Xrs wert foitore rele 1s a forchen af
Lime ¢
a) fD=re AE 2W- Lrerrde =
ay fulore robe, a ee
a*t
—Ne
€
A) fe: ie bo af
—olX —X¢
Rul)= Soe fe (-€@ | Gg"
e
4 foil obey iin ae eee
Get) wate wo
Moke that when f>! (ive rode is an
lacreas/ng fenehon
(61)Terat Probabrtsty Diths booas
Defiarhions
det Xaad Y be alserele ru's Hat font py &
PE D= PlXex, yy)
and rt has He fleeay prope hes! 4. PAY ZO
2. 4% Pog) =4
A] Xand ¥ ate cont cvs Hen ther got deat,
(4) ) Aes He ages Plopetiias :
LGM pire - 2. Li pera dade = i
Zt pag) be te faint pd] of (X14) han
marglok pif of Y 1s pyex)= FPO4)
morginal pu of Yas AiQ)= Fe POW)
4f Xend Y¥ are cond. v's,
Marpined density f x ds Oat re
Morpshed density 4 Yrs ort [Peg rele
Fxamp te
Jojat chensily 4 C4AY) Us ge by
(Op -7 o2xe4y ohgd|
a) Find valve a Cons alee
b) Find marginal deasiwes of 7 Xad Y.
/
aly We elxdly = oft + [Y= Sf ip
IZ x,
=pecat
2 - (62)5) hore fle -loy)| = bik ockel-
[oo 2- ar exer fOr a palf bec ace
L 1
-ftaxdx = [tax |-fax] | Goxau ole = da
3 : ° 8
2 ele | dvedx => vex
ei = i af oeye
{y= 4x = 74 ai fripet ofded
=> Vrs vniferm en thderved (0,1)
Srdepindent Lan dom Veciables
At Xand Y be olscrele rv's, Hag are tadependent d
and only PY) = RO Py):
y
Lf oo i. cont. ry's fheg are shdependent Hand
only 4 (rp Hho dy)
dn the previous example (Daf Oexed vege
and. (OD =p # Ll-tax) =fly ify
=> X and Y ate elepen dent ru's.
Expected Valoe of o fencer a (41)
Let gone bea Junction af (XY) Hen
ELGHYM) =f Lo amg png) dyde where [ery )
Is fount censrty of (x7).
dj Xond Y ave disereke ru's ELGAY = ZZ gen peng
where port) le pdf ef (XY)
(63)Exeretse
het X and ¥ be thdependeat cont. rv's, chow thet
E(KY) = EIXECY)
= recta X,Y are on den?
EC, ,
XY) =f ve x4 (rg) hydlx [exigd= fOr) ly “)
wo
Le Lin ian ge
afar dxf viyg ee
= EX) ECcy).
Covariance
Covariance: between u's Xand ¥
Gv Ke) = E{LX-EMILY- ENT}
Covertlance measures font vert abst hedween
ruls. Xand Y, :
Whin X= 7, CovlxX;X)=ELX-Etx)] = Vasr(X) «
Allcnabive forradla for CoVCHy)
Cov (KY )=ELXY- KEI) - YEO) 4 EQ ECT)
= B(K1) - EO) EC) - EN EC) 4 E(XSECY)
= EUKY)-ECX)E CS)
HY Xand Y ae Urdepea dent ru's ey He exercise abov
CoviKy)=EWMELY)-ECX)ELY) =0
y XY ot Jn dependent ru's Covlx,¥)=0 hud uy
Covlxy)=0, fhi's does not sonpald shdepen dence .
Counter example Us gies below.
(64)Example
&) Find CovtK,¥). b) Are Kand Y jadepen deat ? whq &
a) EX) = ON LF O44) $ 404 DCN g 404420
Mary inet W a, x|-/ oO f
ee re
Morpioal pdf of ¥ to}
4
Ce est
= FOac/) 2 =
x) VEr0+N)2 20 = Ely)
=) Cv(x%y)=0
4) There i's (XY) pare so that PEA) FAOR)
PUX2 0, ¥=0) = 0 # PCX=0), PCY=0) = 2,
Correlation
Corttlabjon coefficient Labor Xand Y 1 hnofed by
Corr (HY) er O,4 and
jet
Sey = ev BY) pe By cy
\darew) \ Vere)
(65)&y Cauchy - Schwartz theguadla'ty 1297 2 4
Coneladion codficient ps measure Ghenr peledfon .
FY Qya-t there Ws pedect nepetive teleddon iY
Pry at Phen Here sk a pifeck positive Ltaor selafp,
: va k Pay 0
y Pe a I \: ae YO 5 Pye
; fe a en Treat
a ee ae
% ae Bees
Example _
At X and ¥ be dependent , exponents
distrr'bu bed cv's wk common pevamder r.
Fiad deasiily oh 2-xX+Y.
X wexplar) , Yarexpla) fen=ae™ : {er aerr
[enpere” nerd
Eay- 0222) =P (x+y 22) = Sf fog dads
2 2x eryee
> Etay= {fren ad, dy
oOo 2-x
= {REMC EN) | ote ° :
oO
= fae BO) Jy
ae ae fe
=f ru fara” oy
HE CE) Pine fax = 6% are*
A aa a phe
=> f02) = Zh ode7*t_xé Ae Nee BNE
D2 a cae = LV Gammn (22, Bat
By Gay 8 AFD)
p@=