Professional Documents
Culture Documents
2. When a card is drawn at random from a p ack of 52 cards and if the card is replaced,
th s
e re ult of the second draw is independent of the first. But if the card is not
replaced then the result of the second d epends on the result of the fir st d raw.
~athelll t' . .
exhausti~ ical (priori) definition of probability : If the outcome of a tnal cons1sts n
e, mutually exclusive, equally possible cases, of which m of th em are
270 ENGINEERIN TY 01sTRIBUTfONS
G MAiHEMA_ p!IOB!_BILI 271
favourable cases to an event E, th~n the pr~bability of the ha . ll~.,, bability of getting (a) a number greater than 2 (b)
3. '[Ire pbro when a 'die' is thrown. an odd number (c) an even
EI usually denoted by P ( E) or simply p 1s defined to be ~~~
ppentng of 1~
number of favourable cases min. "'"1l rr11lll er
f possible cases/outcomes ( n) = 6
That is, P( E) =P = number of possible cases = -m
n r-JurJlber ober of favoUiable outcomes = ( m ) = 4 since 3 4 _ 6
Nuril t ' ' ':>, numbers are
The probability can atmost be equal to 1, because the numb f (a) ble to the even .
favoura
probability of getting a number greater than 2 ( p) = = ~
the number of possible cases can atmost coincide with each ~~~e:avourablecase,
Since 111 cases are favourable to the event, it follows that ( · illJ 6 3
±
fa vourable to the event. n - by
. . This set of unfavourable events is d enoted m ) -£ases
E a,re !lo!
(b Number of favourable out comes ( m) =..,, since 1' 3' 5 are favoura ble to the event.
Therefore
. I
probability of the non happening of the event ~~!yd or£ )
~~ ~by h probability of getting an odd number ( p) = 6 = !2 ~
11-111 - n-m m
q = - - or P( E ) = -- = 1- - = 1_ p (E) 3 1
n
Ill
n n (c) Similarly we have ( P) =
6 2,== since 2, 4, 6 are favourable to the event.
i.e., q =1 - - =1 - p or p+ q= 1
II , Addition theorem of probability
Equivalently P ( E) + P ( E) = 1 11rc probability of_t!1~ happening of one or tile other mutually exclusive events is equal to the
,11111 of the probab1ht1es of the two events.
P is also referred to as the probability of success and th ..
Their sum is always equal to 1. q as e probability of failure 11iat is, if A, B are two mutually exclusive events then,
If P ( E) = 1 , £ is called a sure event & if p ( E) =0, E is called an impossiblemi. p (A OT B) = p (A ) + p (B)
Number of possible (exhaustive) cases ( n) =2 ~ - If A and B are independent events then P ( A and B) = P ( A ) · P ( B)
• umber of favourable cases ( m) = 1
~ Random variables - Introduction
Probability of getting head ( p ) = !!!. = _! In a random ex penmen!,
called . . a real variable is associated with every outcome then it. 1.s
if
funcfi a rahndom variable or stochastic variable. In other words a random variable Ois a
2 Tl .. n 2 11
· from
1e probnb1/ity random t at a. ssigns
on ex · a real number to every sample point in the sample spacendf _a
a pack ,, of
:, gettin
d g (a) k"mg (b) king or queen when a card is drawn at rn dt1JII
0'l 5- Cll/', $. I maybe not~~nmen~. Random va riables are usually denoted by X, Y, Z · · · a it
space s T that different random variables may be associated with the same sample
Number of possible (exhaustive) cases ( n ) = 52 · he set of all real numbers of a random variable X is called the range of X.
(a) Number off ExarnPe I • 1 Wh' l ' · · ted for the
avourable cases ( Ill ) = 4 outcome 'hea , 1e tossing a coin, suppose that the value 1 is associa
d and O for the outcome 'tail' . We have the sample pace
Probability o f getting a king ( p ) = 111 = ~ = 2_ = tH l )-1&:X(Tl"'O
Ran 5 ' T f and if X is the random variable then X( H -
(b) t\umber off n 52 13
. avourabl e cases ( m ) = 4 + 4 = 8 ge of X == I O )
harn
ar· Pie. 2 Su\ , 1 . te two different random
.. ProbabilitYOf getting a k.
' mg or queen ( p ) = -111 8 2
n = -52 == -13
I V
..........
Obie, x ppo ea coin is tos ed twice, w• shall as "'"
' y as follows whm w• have the ,mpJespa<'
7
• ' 7 ~ - - - - - - - - - -~ ~ - - -- - - - - EN
_G_ l-NEERING
= s TRfBUTfONS 273
~
272 ~MAr1-11:AI.\
S= HH , HT , TH , TT
I ~~
, ber of ' head ' in the outcom e. le. z
X = Num txaJlll' . hi of a rticles.
. . f the elements in S to X is as folJows. (a) 1veig . prod u ced by a m achine.
Theassoc1at1on o HH = t - - H T ~~ th of na11s
Outcome - TT "'
(bl Le/lg . h o inter on a s p eed o m e te r I voltme ter.
i,serv1ng t e p
:__.._ _ 2 1
Rnl,dom variable:.. .X 1 (c) O . a s urvey o n th e li fe of e lectric bulbs.
Range of X = 0 , 1 , 2 --~ (d)
Conduchng
ese are some o
f the exa m p les of a continuo us random variable.
. .
SuppDse y = umber of ' tails' in the ou tcom_e_•·- - - . - - - - - - - 11i counting p roblems corresp o nd _to discrete random variables and measuring
c;enerallY d t continuo u s ra nd o m vanables.
011I come _ _.J_j_ _H_ H_ _-r__H_ T_ _--t-- -..::
T:....:H
c..:___==---·- ~ hl~SN 1 O .
pro . the category of the ra ndo m va riable we have two types of probability
- Random variable X _,_ 0 1 _:~ According 10
dJStributions.
Range of Y = 0 , 1, 2} L__ - -
• Probability function and Discrete probability distribution
Example - 3 Let the random exp e riment be throwing a pair o f 'die' and th
pace 5 associated with it is the e t of all pairs o f numbers cho en from 1 t~ ~ If for each value •r ; of a discrete random variable X, we assign a real number
p(x,) such that
1l1at is, 5 = { ( x, y) I x, y belong the set of nos. 1 , 2 , 3, 4, 5, 6 } (2) L p ( X;) =1
To each outcome ( x, y) of S let us associa te a ra nd om variable X = x+ y
then the function p( x) is called a probability f11nctio11.
The range of X = 2 , 3, 4, ··· 12 co r res p o nding to th e 36 outcomesinS
namely ( 1 , 1 ) , ( 1 , 2 ) • • • ( 1 , 6 ) , ( 2 , 1 ) . . . ( 6 , 6 ) If the probabili ty that X take t h e va lues xi is P; , then
We have X ( 1 , 1 ) =2 ; X ( 1, 2 ) =3 = X ( 2, J) ; P( X = xi) = P; or p ( X; ) .
X ( 1 , 3) = 4 = X ( 3, 1 ) = X ( 2 , 2 ) e tc. The set of values [ , . , p( x ) ] i ca lled a discrete ( finite) probability distrib11tio11
1
~- ~ Discrete and Continuous random variables - Definitions of lhc discrete ran~om v~ria b le X . The function P ( X) is called lhe probability
nd density fimctiou ( p • d. f) or t he probabi l ity mass f1111ctio11 ( P · 111 · f)
If ~~a om variable takes finite or counta bly in fi nite numbe r of va lues then itiscalll'II
a ' f:;ete rnu do_m variable. Here coun tably in finite m ean s a sequence of re~ Thedi5lribution function f ( x) defin ed by
~~~-1 ers.range.
1 It is evident that a discrete ra ndo m va riable w iJI have fi nite or countabfy
X
f( :i.) == P(X s x ) = L p(x;), x beinganinteger iscalledthe
If a random variable tak · · ailed a
11011 discrete O '. es non countable infinite num ber of va lues then it 15 c 'ft/II i = I
range of a rand' coutz~mous random v ariable. Equ iva le ntly we can say that: 1ou; CU11111/11 tive d. .
om variable X · · X · contrnu zstr1butio11 f 11nctio11 ( c . d • f)
random variable A . is an mte rval o f real numbers the n 1s a . tel'I'~
- and variance ( if" l for a
of rca I number . • continuous ra 11 d o m varia
· ble can assume any va Jue in the lil
f
\Ve arc air d
&eneraJ ea Y fa milia r w ith t h e d e fini tions of mean ( X)
Example-] reque ncy distri b u tio n . T h a t is
2 lll _~-
,.,
-----
t{,'fl<'.t' P ( ··- -- - - - - - - --
21.
11
I(
1111111
11t•r of /t'lt-/1l111111· 1111..,_ t,11,11
- --
3. :- ., :.· -::,1· :·:, '. !/,,,!'111 • d1-•r1l•ut1<"I r,pr,·,,·m- 1 ,//,, ,,·r pr / ,1il'/1tv di-tllbl r l\dl.l\ t.' P( \ l = "t· /'\ q" \ = IOc ( 0. 1 )' ( 0.9 )IO-t
\ l
~ " "'.:. .•: .;n_i: 1rr,u:u·
10
Ill (l (ii Probab1hl) th.i t n o lme is bu y = P ( 0) = ( 0.9 i = 0.3487
.. \ 1
(iii Prob,1b1ht • th,11 all lin e~ ,-.re bu ,. ; I' t JO l = ( 0.1 )to
Gii) Probability th,1t a tlc;i..,t o ne lme 1-. bu-.,
» •'ll,t---en,•lhat ."(, 1 '> l) t o r ,l II \ .inu'~.:., 1• 1 \ l
= Probability of no line 1<; bu::.\
\teJn =u = ~:• \ l \ = l l).,.t,(.)-<)(l.,..H) =- :!5
- 1- P ( 0 l =I- 0.3.J '7 = 0.6513
\ ananc,' = ·. = ~ I \ - µ I~ I' I \ )
(ii )
Probabtli t~ that a t most 2 lim...., ,,re but,\
= P(O)+P ( l ) +P! 2 J
9 2 1
= (0 9 ) 10 + 10c ( 0 . 1 )1 ( 0.9 ) 'T !Oc ( tll 1 I L o '
I ,
Thu, = 0.9298
~-----..:..___:_:_= ~--- --------
4. FmJ
\tean = 25 and Variance= 75
1 10 10 ~ ~ -"d~tandarddeuzatwnoithenumberofcomctlyafbuwedmust
P(:t 2: 6) = JO
2 [ c6 +lOc7 +10c8 + c 9 +lOc ] fftrJmLa,1a; ,:denbare25and '\1875 ~'.ndanest,mattofthenumbtr;:,:;~t~t
JO :i 411 t~409 5 ectly (i) 8 or more questions (11) 2 or /es, (iii) 5 qutc;t,ons , I@ es
1 ~ F~~~ --
= (210+120+45+10+ 1 ) = = 0 377 ar~ n (µJ = np and 5.D( cr ) = 'inpq forabmonuald i tnbution.
10 1024 · . h,1\·e rnea
2 77 \\e q = ✓1.875 or n p q = l. TS
_ nd ..r;;;;p
- 2.:> a
Thus p ( x 2: 6) = 0.377 51· data "p - · - q = 1.875 : . q = 0.75 ; p = 1 - q = 0.25
. have 2..:>
In the <,CCond case when there are 4 options t1ence1,e have ( 0 25) n = 2.5 :. n = 10
- 25 we .
p = ] / 4 , q = 3/ 4 ; II = 10 11p -
S!f1Ce h umbe r of correctly answered questtons.
denote t en
[,ell .,.r II - :i. = 10 ( 1 / 4 )x ( 3/ .q10 - r
P(x) = 10c ( l / 4)r(3/4)10-x = ~o [ 310 -x· lOc ] P ( x) = nc r q cr
l 4 l l
Hence P ( x 2: 6 ) = P ( 6 ) + P ( 7 ) + P ( 8 ) + P ( 9 ) + P ( 10 )
It,
P(X) = 4~0 [rncl ( 3)10 -r ]
1 [ 4 3 2
= 10 3 • lOc +3 10c +3 lOc +3 · 10c +10 ] . te is needed for 4096 tudents we ha\e
4 6 7 8 9 c,o SIIICe the estuna
p
] 4096
)10-rj-~[10 (3)10-r]
= 410 [ 81 210+27 120 +9 45+3 10+1 ]=0.019 4096 Pc x) = - - [ 10c ( 3 - 20 er
X X X X 410 X 2
23. In ,a111pli11g a large number of pnrts 111n1111fnct11red by n compnny, fire mean 1111111ber11 (i) We have to find f ( 8 ) + J ( 9 ) + .f ( 10 )
defrct,ve, 111 ~ample, of 20 is 2 011/ of 1000 suclt snmples how mnny would beexpertd
t,, co11/n111 nt/enst 3 defective pnrls. = 1 l 10 . 32 + lOc . 3 + 1 I
256 CK 9
Mean ( µ) = 11 p = 2 by d a ta, where n = 20
I C, 20 fl = 2 1 I 9 . 10 + 3 10 +1I = : c -t36> = t.703 .. 2
:. P = 1110 = 0.1 Hence q = 1 - p = 0.9 256 c2 c, 256
Let t denote the defective part. . 8 or mo re questions i 2.
umber of students correctl y answenng
p (' ) = nc if f/ 11
-
1
= 20 ( 0.1 ) 1 ( 0.9 )20 - ' (ii) \Ve have to find f ( 2 ) +f ( 1 ) +/ (0 )
l (l
1 9 10 I
= [ 10 38 + 10 +3
l'robabtl1ty of atlca<,t 3 def 1·
< cc l VC part!> 256 c2 c, · 3
8
= P(3)+P(4)+ · ·+ P(20) 3 3 1~2 8 .. 2I 53
= 256 ( 45 + 3() + 9 ) = 256 ( 84 ) = 2 ::, . . ~
= l - j P(O)+P( I )+/J(2) J
No of 2 uestions ss 21:,3.
· Sludents correctly ans wering 2 or less than q
= I- I ( 0.9 ,211 + 20(' ' 0. ) ) ( 0.9 ) 19 J 20 . ( 0. ) >2 ( 0.9 ) IM I (iii) W h
I (2 e ave to find J( 5 )
0.12.1 1
I hu, lh1• numb ==- I 10 3,; I = 239.2 .. 239
N 256 c,
"r of d1·f<,r tivt><, in l(}(J() 'it1mpf,,., j ., IOOO x 0.323 = 323 umber of 5 t d . 5 estions is 239
u ents correctl y a n s wenng qu
296 ENGINEER 0~11,s:::r.:Rl::B:.:U_T,_o_N_
s _ _ _ _ _ _ _ _ _ _ _ _ _ __
------------------------=.:.:..:::.:::.'.:::_:1!"1N,G MArH
E~r/Cs ~tJTY -
25. A11 air /me knaws that 5 % of the people making reservatio11s O
. . .
~
II a certntnjl'1 h ' have to find f ( 0 )
turn up Consequently tI1e1r pofzCJJ 1s to sell 52 tickets for a flif!.I t 11 /s t Un// ,:1 \\ e
tJJJ
-
Sc = 2.-, x 1 = 25-
people. Wiza/ is the probability that there will be a ·eat for every p~ zat can only hold; f( O)= 25 o
sengerwhot ,
4
» The probability ( p ) that a passenger will n o t turn up is "'1ls p> ....ber of families with 5 girls is 25.
q,ecied nu .. ,
,, = 0.05 :. q = 0.95 [ , have to find f ( 2 ) + f ( 3 )
Let x denote the numbe r o f passengers who w ill not turn u p . 1iiil \\e _ - _
- 5 +25 · _.,c = :>O Sc = so x 10 = 500
P ( x) = nc JI q" - .t where 11 = 52 = 25 c J2 2
l
S1m:e we ha ve to find the <..'xpectcd n umber in r~pect of 800 fami lies we haW
lhts lhe binomial probabiHt)' distribu tion
function. , t ,m
bt ain.:-u r(
Sc ~
1
• •
d1·
< , d frequenc1,a are 0
r q::: ~e Were th rovvn 96 time , expec~e
&XJ P <x) = 800 32 = 25 Sc = f ( x) <~ay) 96 ( ' ) whe re x = O, 1 , 2 , 3 , 4 , .,
X
(i J We have lo find ( 3)
~er f ( x ) == 96 5C = 3 .5C
f < 3 1 = 25 :;c = 25 x 10 = 250 25 .l T
3
· - 6.76;:, == 7
F(l) - J004 We have to f d 11
- · C1 ( 049 - )3 = 400 ( 0.49) ( 0.51 )3 = 25.999 = 26 rn <; u ch that
. ) ( 0.:,1
I C., f' ( y ~ 1 ) > 3 / 4
F(2)= 100"'c 2
2 ( 0.49 ) ( O 51 2
.
. ) = 600 ( 0.49 >2 ( 0.51 )2 = 37.-17 = 3'I ic l-P( ,< 1) >3/ -l or1-P(0 )> 3 •
F(3) = 100 4 lu 1 - ( 2/3 II 11
. c , ( 0.49 )1 ( 0 - ?~ ••e ca 11 n d ) > 3/ -l or (2/3) < 1/-1
· .:,1 ) = 400 ( 0.49 )3 ( 0.51 ) = 24.()()04 " . n n b Y in!>pc
· tion a~ we have
F( 4 ) = J004C 4
Th . 4 (0.49 ) - 100( ' 2 3
us the .
required th
- 0.49 )• = - 765 - 6
:,_ -
t.
01 ence th/ "' 0.67 ( 2/3 i-., == 0.--1--l , ( 213 >3 = 0.3 , ( 2/J f' = 0.2
eoreticaJ fre u .
q e ncies are 7 , 26 , 37, 24 , 6 ~ ed
_.:.:._~------ ---
11 is 4.
't-.l! !1J
;; t - H = t - l ';;i?t.r
~ - ------
.,---=-3-~-4~ -;----;--
- ~ .-! .r C o
- r+l .x! (r-.r pp -p If
r li3 168 '57 IS 3
Pt~-rl) = ~ p •! i,t-r ~ \',e b.i,e for the Poisson dismbuticn:
:c+I q.r!( ll-z)r P
"i.f x o+ 168 + 74 -t 54 + 12 + 5 = ll.J'8'l5
- ~ e
-.r+ 1 - ,,
"'c;r P'"',P-I
'f " }lean ( µ ) = m = I.I = 400
By llSing (1) in the RAS -han;. ,,/'e-"'
The Poisson distribution is p ( X ) = - - ,-
a-.r p :r.
P ( .r+l) = -- - P ( .r)
.r+J q let f{;r) = 400P (x)
( 0.7825)"' e-0.7825
But
r = 400 r!
0 2 3
' 122 60 15 2 f( r ) = 182 9 ( 0.782S f
» We shall first compute the mean ( µ ) of the giwn distribution. . .x! . . =0,1, 2.3, 4.Sin /(1 ) il1111P!'f
lheoteticaJ frequencies are got by substituting x
I_fr O + 60 + 30 + 6 + 4 are as foUo"
µ:If= 200 =0.5 . ) 1 = 183 07825 )• 1C
( 182.9 ( 1829) ( ·
We have mean ( µ ) = m for the Poisson distnbotion.
The Poisson distnbution is ( 182.9) ( 0.7825)2 = 56 <1829 ><6o.7825 i. 1s
2
~ t-m
P(r) =-:;-;-- and letf(.r) =200 - P(x) ( 0.7825 )4 ( 1829 ) ( 0:rt/15>5
- • 0
( 182.9) - - - • 3 120
( 0.5 )"' t- 0.5
f z J = 200 - ~ - - But e-0.5 • 0.6065. ~ th 24 15 3 0
r! tbe eoreticill bequencin are 18.l, 1~ • 56 ' ' '
~ sTR,sur,o,,s 303
302 ENGINEERING MA
THE~Tics
~ fses
i~11a111ifact11red by afirm arefound to be defectivt Find tire p ob bi/
34. In a certain factory t11ming out razor blades there is a small p r o b a b i ~ ,o/ooftire ~ :,,x 2D0fi1se:; contains r a ity that
blade to be defect we. T1ie blades are supplied in packets of 10. Use Poiss d' "'1\/ for qny J6. - . co11tam1 .. d ,r. .
1stn obin ,r. tivcfiises (11 ) 3 or more e1ect1ve fuses.
calculate tl1eapproximate number ofpackets containing (i) 110 defective~?.) °ution io (i) ,ro df)ec .
11
(iii) two defective blades in a consignment of 10,000 packets. onedeftctr.y b'lity of a defectzve fuse = 2/100 = 0.02
" pro ba '
» p = probability of a defective blade = 1/500 = 0.002 17
P
mean nu
mber of defectives µ = 111 = 11 p = 200 x 0.02 = 4
In a packet of 10, the mean numbe r of defective b lades is mze- m
111 = II , , = IO X 0.002 = 0.02 . distributio n is given by p ( x) = - -, -
n? e- m e- O.Ol ( 0.02 )r The poiSSOn x.
Poisson distribution is P ( x) = - - - = - - ~ --'-- T - 4
x! X !
4 e But e- 4 = 0.0183
Let f ( x) = 10,000 P ( x) ; Also e- 0·02 = 0.9802 r.e., P(x) = ~
3 4 + .!. m 6 or 1 3 2
4 m +4
1 4
m
'I ·~ "'--m
2
-- 8
8
=
mX e-m i.e., 2 2
» Wehave P(x) =
x!
and P ( 3 ) = P ( 4 ) by data.
4 3
,n2 _4 == O or 2
(m - 1 ) ( m +4) =0
/11 +
m3 e- m m4 e-4 i.e., +1 m=±2i
Ill ::= -
- --- - - - I
f11nctio11 ( p.d.f)
Mean
- f x·f(x)dx
(µ ) - ... (5)
If a b) i a subinterval of the range space of X then the probability that x . - 00
f /(x) dx Thus
P(x~r)=
r
P(:r < r) = 1-P(x ~ r)
. .~ I I 1<x>dx= 1
µ = r/o-~co-1) ] =~ ; µ=~
Eddently f ( x ) ~ O
x-µ x = µ +✓
2 cr t , we have dx = ,5 cr dt
a2 = a f (x- µ ,2 e-ardx Putting t = r.:-
v2 cr
or
0
t also ,·aries from - 00 to oo
Appl~ing Bernoulli' rule we have,
~ 00
= a
- I
-
a
~
0 - µ- · -
2
2
re
?
0 - ( - µ ) . _ -=..
a3
0- 1,] Lt.,
1
But .u = -a But
J e- t2 dt = -"7c2 by gamma functions.
0
1
H ence cr = -a
f J(x)dx=¼ · 'In =1
-oo '\i 7t 2
Thus for the exponential d istribution
1hus / ( x ) represents a probability density funcl!on
. .
.Mean ( µ ) = ..!..
a '. S .D ( CJ J-- .\Tote: N . . .. . /,11wm1al dl:'tnbutum u!tt 11
a n is I onna/ d1s_tribution can be derived as a lrm1t111g ca,-e of the .• J, l ,;:: r. 4 "° and
arge • \\' 111 derrt, v \
Remark : Wean - S Dfi P::: 11 .,,,_' ; iezther p nor q 1s i 1er11 ~mall t ,.
- - - - ·__-_ _-~.=_0 ' ~th:::.t_:expo
~'.:.n:e~nt1a~l~d~1s~t:n~ ~
·b~u~ti~on~.: __ _ _ _ _ _ 2
oisTRIBUT/O,..
N::S_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _:3:..:1:.:.3
~
312
:-----------------------===_::1!TH£1..f.,q71cs ENGINEERING MA
2
_~ f 2d2 t2 e-
00
= cr &
x -µ
Putting I = Ti;; or x = µ + V2 <J I , w e have d x = ..fia dt
t also varies from - oo to oo
= J!.. -f _,2 ri
V7t __ e dt + <J --fir
-I t e- 12
dt
Mean 2 µ -J _ , 2
=-:;rrr e dt +
I ..
= "Ji; f ( x- µ )2 e-(x - µ J212a2 x=µ X
-- dx
Substi·tuting I = ~
1'1,e area ~ ~I µ . divides the total area u nder the c11rve wilicli is equnl to 1 in to two equnl pnrts.
1'1,e line
V2 <J we have as . th
in e ea rl ie r ca se,
1e rig ht ns we I1 as to the left of the line x = µ is 0.5
d
314 ...., DISTRIBUTIONS 315
~e1Lf1• '
• Standard N orm al Distribution • the Rf-15 of (2) geometrically represents the area bounded by the
;ntcgn11 al c u rve r ( z ) bctw,•t·n z = z1 and z = -½· Further in particu lar if
111
'fhC
/1 dard ,,orJTl
Wl• have /J ( 11 S x S /1) = f j( x ) dx 5,an
, 0 we hJVC
=1 , z
ti
The integral in the J~ I IS of this equation cannot be evn luatPd by known ml!th <l
inl egrn liun and wt• hnve to employ the technique of n umerical integration° ~-of
becomes tedious. I le n cc w e think of stan da rdizntion and the same ii-. as followb~v tch
l'ulting z = x-
-~µ or x = µ+oz we h ave rlx = adz =
Ll'I Z-i -cr
= n - µ and Z2 = b-0-
- µ be the values of Z cor rl'spond ing to x = n and I ¢ ( z J a Iso. d c no tc Cl by A (
z ) rep resents the aren (shaded portion) a~ ~hown in the
. . . _
x = b. The intcgrnl in (1) nssu mes the following form . figure. Since the to tal a rl'il b 1, the a rea on either s id e of z = 0 I'> 0.1.
z2
T bulated va lues whic h gives th e a rea for d ifferent positive valu~.., of z ar1· ,,· 1i!ablc
I ' I a S xS I, ) = d- f e z 12 d z
'V2 7t
2
a~d thi~ helps ui-. in p ra c tica l problems . T he procedure for u~ing the tar,, w, 11 be
z, discussed Inter.
.,
Table is given at the end of tile book I
2
H c n cl', !' ( a S x S b ) = I' ( z1 S z S z2 ) = ✓i. ii. z /2 dz ", (2) WO RKED PROBLEMS
45. Find which of the followin:.?, f unctions is n probability dcm,it y function
J 2
If F ( z ) = fu e- z 12 (stnndnrd normal probability den.~ity Jun ction) , it may be (i) fl ( X ) = { 2X, () < X < 1
0 o l/J erw i~ e
observed th.it this is same a"> the p.d.f of the normnl db lribution wi th µ = 0 and
<J = I. '111118 we cnn H ny that the normal probability density Junction with µ "' O 0nd (ii) f2 ( x ) = J\ 02x, - 'I < x < 1
<J "' 1 iii lite standard normal probability density f unclion. z = ~ ~ µ is called the ,,;ta ~'
11d d othen11 ise
normal vn~iate <S.N. VJ nnd r ( z J i.~ en/led the standard nnrmol curve which is symmetrical (iii) fJ ( X ) = { ~X I, Ix I S I
about the /me· z = 0. 17,e curve is os follows,
rp (z) (' !
IV) /4( X ) = 2x,
4 - 4x,
()
ot/1erw ise
0 < xS 1
I < X < 2
otherwise
>> Cond 1·t ions
· fo r n p. d.f n rc f ( x) 2: 0 and Jf ( x ) d x --
z
I
316
srRtBUTtONS_ _ __ _ _ _ _ _ _ __ __ _ _ _ _ _ _ _317
_
~~~=~:-:-~- :--- - - - ---------~E~N~G'._'.:IN~E=E~R~IN~G'. _lvl~A~T:HEA.l.'1r,cs."'
(i) C l ea rly J ( x ) :1?:
~
l
0 x
-6 +c,
r
I c such //mt f ( x is a p.df
f f, ( x ) == f /1 ( x) d x == f 2 X d~ =[x 2
= Jc. f llld
t/le va/11e oif
)
0 elsewheree
"" 0
0 0
/i ( x ) is a p .d.f
p ( J~ x ~ 2 )
( ii) • I Msoft11d
f° )c give n function ca n be written in t he fo rm.
Also we must have J f ( x } dx = 1
l ,, f( ;( )~o if
2x , - 1 <x< O c ?: 0
f2 (X) = 2 x, 0 <x< I
0 o therwise
In - I < X < 0, / 2 ( x ) = 2x
1
is less than 7.cro.
II ,
J (i + c)dx =
Fu rthe r J/ 2 ( x) d x = f 2x dx = [ x2 ]
1
= o
0
I J
a[~+;,!
-[;(', ·I X:[ 2 t- 2 ...;
=
1
f(4 t2 J - ( 1+1J f ..:. ,
I
12
f,., ( X ) ; .. it p.d.f
(i v) Tlw r,1 v1•n furn li1111
f1 ( A'.) 2x ,, ,, ;,, () ,, X "': /
-
'fhu~
/' ('J ~ X :::: 2) J/3
I
318 ENGINEERING 01sTR/BUTfO..N-~
s_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __:3~1:9
·
3
f A1 2 d.1=1
~ ~ -
-'">-< 1 )(e- ·')T
1
=1
,.r., k [ x(-C lJ
i.e,
0
or 9J.. =1 k =1
9
,, , k [-; - u-Ill = J
k(1-!)=l
(i) P ( l < x < 2 ) = !/( x) . , = !~ d, = [ :, I = :,
,e,
e
k = --:,
e- -
1
(ii) p ( .:t S I ) = f
o
f ( x ) dJ. = f x2 dx = [ x3 I1 = 1 Mean == µ = f x · f(x)d.:t= f
0
x - -e- xe- l dx
e- 2
O 9 27 27
3 3
f f
(iii) p ( x > 1 ) = f ( x ) dx =
~dx= [iI= !:
(iv) Mean =µ = Jx . f ( x) dx = f
0
x . x2
9
dx = [ x-!
36
I = 81 = 9
36 4
[_2__ 2_
µ =- e
e-2 e e
2(2__
e
1)]
(v) Variance V = f :?j(x)d.:t-( µ /
µ= -e- 2 - -5 ] = -
2e-5
-
e-2 e e- 2
x2
3
0
f x2 - 9
d:,. - (2)
4
2
49. Is the following f unction a densihJ f11 nctio11 7
> 0
l
-r
=
V =-
L 45
{I _ 81 - 81 81 27
16 - 15 - -16 = 240 - 80
f ( X) e
0,
,
X
X -
3 F (x ) = J ( 6x-6x2)dx = [ 3x2 - 2x
3
J: =
3
x2- ix3
That is, f k x2 dx = 1 0
- 3
c.d.f = 3x2 - zx3 if O S X S 1
1 X
k= -
or [k: 13= 1 18 (ii) F (x ) = J ~4 e-x/2 dx
0
12
(ii) p (x S 2 ) =_J J_I x2 dx =_]_ rx3 ] =_.!_5-l ( 8 + 27 ) = 5435
1 _3 _
2
= ¼[ - 2 ( x e- x/2 - 0) - 4 ( e- " - 1 ) ]
b
~--
322
- -------- - - - - ~ ENGINEERING .\(A.,.,, o,sr,R~:::'a:.:u_r_,o_N_s___ _____ _________ _ __~
the p.d.f of the e,ponential distribution. ~ .. 2
F(x) =
\
_f f(x) d.,
-=
=
.\
0
J a t•-a'dx = - [ e- a , r0
= 1 - e- a x i.f., J
0
k X t' 4 ' d,· --
----
c.d.f=l -t'- a .1· if O< x < oo putting 4 ·'
l I=
Q , X $ 1 k
F(x ) = c(x - 1) , 1 s x
4
s 3 Fi11d c 1111d 11/,:0 t/,e p dj I e.,
: [ - l '- 1 1 or
8= 1 k =B
1, X > 3
54_ A rnudo1t1 varinlllc x /ms tl,c density f 1111ctio11 f ( x) = k / 1 + ,!- , _"" < l <
» \Ve know that the p .d.f /( x) = -d [ f ( .r) ] Determi11c k 1111d ltc11cc cvnluntc (i) P ( x ~ 0) (ii ) P ( 0 < x < I )
00
dx
0, x $ l » We must have Jf ( x ) dx =
1
f( x ) = -lc(x-1}3 , 1SxS3
0, > 3 -
X
I(.,
I- - I+k~
y-
d., = 1.
f (x ) ~ 0 for c ~ 0 and we m ust have JJ ( x ) d:r = Since the integran d b even we h ave ,
3
53
P(x ~ O ) == J / (x) d .,
. Find Ir '° rhar the ollCJW1mlfunction can seroe n:. a probability density Junction oJar(1114£"' 0
vanabie.
1 1
_r ( x ) = , 0 , for x $ O ==
Jt I -I +x2
- - dx
0
k x e- 4 x- for x > O ~
7
== r tan- i \:
=
It
= ,
n Jo
» \\ e must have J f ( .x J dx = 1
p (X ~ 0 ) :: 1/2
c. 7t L '.?_ J -
324 325
s -- - - - - - -- - - - - -- -- -..:.::.:.
CJ4STRIBLITION
" '::-
25 )
I..II> To find P ( ·' $
2.,i;
l
:r f l I l I - I ) dt = P ( ., S 25 )= J
0
l 1 ,( = [ - ,·- , / ~ = -t'- 5/ + 1
t" t
It A.
:> 3 -0
= or >.- 1
Lo
1
]= I k = b
P ( x S 25 ) = 0.4&47
P O b1hn !hat !ht> pn>jt-\.1 will tit.- rnn1plnt..,I
Pl"Ob.l (iii) To find P ( 16 S x S 32 )
I < t < I 3 1 ~In(<'
. c 1. • rn ye.ir-,. ' ' Ill • mont h,- j.; t-.:1ui,-.11tnt to Iii:
32
d.x
p I I) < t < I ' 3
I 3
J b I ( I - t ) dt P ( 16 s .\' S 32 ) = J 401 t - 1/ 40
)= 16
0 -x/40 iJ2 - - - -l :- ., ,-z. 5
= [ -e .he - r
= ~ 3 12 - 2 ,1 l l/ 1
= [
I 2 l 7
Thus the r · ..0 .1 _,
-,- t=
_,
-,- P(l6 $ x ~ 32 ) = 0.221
equ1red probability is 7/ 27
56. . van·atr wi·t h mean •, find ( i) Pl -., l \ (irJ P \ t .._ 3 I
~ Wolttdrt run <m 1/wu:.a,ub 0 57• I'l x 1s an t x po 11c11t1al 1
(16'1'
Uthiclt h Q rand(,,,, t'CI bl · / k.m J w1thmit a1111 ,-ort ofprnblrm 111 ,rsptcta/• >> The p.d .f of th e e xponential d istribution I'- ~i,cn b\
r1a •· havmx pdI
l
t: X 4fJ
ftx 1 .a , 4tJ l~O (l (' O.l O< .\ <oo
/ ( X) = ,
\
0 o th l•~;..,._,
(J
f .l: ( ()
litd th,- /Jr,.1..,L The mean o f th i-. distrib u tion is ~1,·,•n b\. l •u
"""1!/i,1/ lhat th
' v,:Ju, /p b lr111,I,/,. /Tf't' By d ata, lnt.>an = 1/ u "' ., . (l = \ 3
01sTRiBLJTfONS 327
326
~ 0 10
0 =0 + J f ( x) dx = f .!_5 e- x/S dx
1 1
P(x > 1 ) = e-V3 = 0.7165
3 P(xSOor x::?1) = - [ e-x/ SI = - (O - e- 02 ) = e- 0-2
(ii) P(x<3) = f f(x)dx
0
p ( x s; 0 or x ~ 1) = 0.8187
3
59. The length of telephone conversation in a booth has been an exponential distribution and
J .!. e-x/3 dx found on an average to be 5 minutes. Find the probability that a random call made from
0 3 this booth (i) ends less than 5 minutes (ii) between 5 and 10 minutes.
= - [ e- x/ J r o
= 1 - .!_
e
= 0.6321 » We have J(x)
By data 1/ a = 5
= ex e-=, x > O ; Mean= 1/ a
ex= 1/5
p ( X < 3 ) = 0.6321 Hence f(x) =~ e- x/5 1.s the p ...
d f
5
5
58. If xis an exponential variate with mean 5, evaluate.
(i) P (O < x < 1) (1·1·) P(
(i)
P( x < 5 ) = J J(x) dx
-
00
< X < 10 ) (iii) p (X ~ 0 OT X ~ I) 0
J ~5 e-x/ 5 dx = - [ e-x/ 5!
5
»/(x)=ae-<Uo<
, x < oo ; Mean = -ex1 = 5 .·. ex=
1
s =
Hence /(x) = .!_ e- xlS 0
5 , D<x<oo = 1 - (lie) = 0 .6321
1 p (X < 5 ) = 0.6321
(i)
P(O <x< J) = f f( X) dx (ii) 10 10
0 p(5 < X < IO) =
J J( X) dx = J -15 e- x/ 5 d
X
= f 1 I
-5 e-x s dX -- -
[ e-x/ 511 P(5 < X
5
p (0 < 0
X < 1) =: 1 - - 1/5
P(o < x < e == I-e-0.2 = 0.1813 .-. P(s = ( 1/ e) - ( 11e2) = 0.2325
l ) = 0,1813 ~ x <_1:::._:0~ ):_-:=_:0~
.2~32~5:__ _ _ _ _ _ _ _ _ _ _ _ __
..
ENGINEERING
~, ~.• ~ rt
C)IS TRIBUTIONS
329
~
328
J fl! a..,.,,, "
T!:(:'-ll," ,~·• ilY •• ~
.L ,
1 • ,., , ,11enti111J11
:n,;, . . ..,_ 1 · .11,-.trih,u,i u•rtl: tire· ~-r,.
..,, ..:,.:r !'Ill
Fi 11d tlit· l'n,l,.il•rlrty t /1111 ti,,- 11ct l'"-'fit . ' r o~Jt:
10
Pr.-.t>ar 1lit1 ,,t r n.,fit .., ,-.., -Jmg <~) L'- l'qU,11 t,, Worked Problems on Normal Distribution
l - fu>t- 1pn,11t ~ lh.~ l )
repn.'S<'nts th.- ,u,·.1 un,l,•r the :,tandard n o m,al cu n ·e from O to :. Tat.>u ut~ , a.h.:es
_ J \0 .tll k' O.ll h d., = l +[ <' ti.Oh ~ ;- e,• 3.75 " lu.:h ,,,...,, the ar,•.1 ior d ifferent positive valu ~"' of : L, rt!ad 1lv a, J.1la bl..-. Su,.h .l t.1r 1e
,, J.l\ll°l?,ous to th,• formal of a lo g arithmic table i:, ,.ill~ n,>rnra, rr-t'..h l:•v ::.ci.· \ \ e
r n'><'llt a few illustr,1t11..1n-. g<'<)ml'trically, theoretically and \\nt..- th..• , alue b, m.1ku,
u.,.. of th,• t;ibk· ( s n·,·11 u l tlr,· ••tul ,,f tlr,· 1,.,._ ). \\'e a l~ b..-,u m mmd the tollow m g
~ rn>b.lt-th1:1 th.1t profit ,A ·,, -d:, R.... : l cm .1 ,-.mf_k• d,w 1,; ,...u l>li h....._1 r,,u lts
Thu, th.- rroo.il>ility thJ.t 1t ~ f'<',1~ ,>n the tollow in ~ d ,w i-< 0
O 1.76
- _1=
-&- 0.44
f e-z212 dz+ _&1_ f e-z 12 dz 2
0
z
1.55
l
Theoretically the area = r;:- f
e-z212dz = cl> ( 1.55) 1
= --
0.44
f 2 1
e-z/2dz+r::=-=
1.76
f e-z l 2 dz
2
'12 7t 0 & '\/27t
0 0
Referring to the table we move vertically down along the column of to re ch = cl> ( 0.44) + cl> ( 1.76)
2
and then move horizontally along this line to the value 5 ( regarded as .05) to ;t L5
with the numerical figure 0.4394 ersea == 0.1700 + 0.4608 == 0.6308
Hence ¢ ( 1.55 ) = 0.4394 Equivalently P ( - 0.44 S z $ 1 .76) == 0.6308
Equivalently we have P ( 0 S z S 1.55 ) = cl> ( 1.55 ) == 0.4394
Illustration - 4
Illustration - 2
Area of the standard normal curve between z = 0.57 to z == 2.49
Area of the standard normal curve between z == - 0.86 and z == 0
cp(z)
- .86 0 z
1 of 0.86 =0 =0
Area = -:-t::=
i
e-z 12 d2 _ 1 J _ 12 2
Required area = ( Area between z to 2.49 ) - ( Area between z to 0.57 )
-v2 1t- 0.86 - _ r=--=
'12 7t e z dz by symmetry. Required area == cl> ( 2.49 ) - cl> ( 0.57 )
0
:. . the required area == cp ( 0.86 ) = 0.3051 == 0.4936 - 0.2157 == 0.2779
Equivalently p ( _ O.B6 < Equivalently P ( 0.57 S z s; 2.49 ) == 0.2779
IIIustr ti - z 5 0) = 0.3051 Illustration - 5
a on- 3
Area of the stand Area of th
ard normal curve between 1 76 e s tandard n ormal curve to the right of z = - 1.96
z = - 0.44 and z = .
q, (z)
srR/BUTIONS 333
~B/LlrY DI . . .
v nrinte with 111cn11 30 and standard dev1at 1on 5 find th_._._,l't
. i< a 110/'IIIII1 ~ I y I1rat
,etWl'Cll:: = - 1.96 to O ) + ( Ar~a to the . h 6-1, lf.t . (ii) X ~ 45
- ( An•.i L rig t f .,., ,, S 40
R '-luin.~j
c
,lf'l.',1
= (Art'il lwt\\'l-cn : = 0 ,lllli 1.96) + 0.5, by symnwtry. o l • 01
- (i) 26 ;;:, ..
x- µ x- 30
= cp ( 1.96) + 0.5 = 0.-1750 + 0.5 = 0.9750 tandnrd normal var iate (s.n .v) z = = --
p ( _ 2 _ I%) = 0.975
,, We i,avc s cr 5
Equivalent I)' - . . (' d p ( 26 S X S 40 )
(ii 10 in • -
_ Ei'fllunt, th,·_ft,/l,,wi11s 1irc>l>i1bi/iti~ with tire hell' of 11nrmnl probability~
6 z = _ 0.8 , If x - 40 , z = 2
62 lf ,t= 2 ,
(i) p (: 2 0.65) (ii) P ( - 1.64 S :: S - 0.88) eed to find P ( - 0.8 S z S 2 )
Hl•ncl' we n
(iii) pl:: s - 2.43) (iv) P ( I z I s 1.94 ) p ( _ 0.8 s z S 2 ) = P ( - 0.8 S z S O) + P ( 0 S z S 2 )
» (i) P ( :: 2 0. :; ) = P ( :: 2 0 ) - P ( :: S 0.85 )
= P ( 0 S z S 0 .8 ) + P ( 0 S z S 2)
= 0.5 - $ ( 1) = 0.5 - 0.3413 = 0.1587 P(x < 1950) = P(z < - 0.83 )
d, 1 ·b · d
= 100() X 0.6826 = 682.6 = 683
1 4 rm P11rclrn es 2500 b lb Ji d
{i) more , 11 s m tire n11111ber of bulbs tltnt are like I!J IO fas/ far
, iooli¢ >> Let µ and o be the mean and S.D of the normal di5lribution
11,a,, -lOO l,orirs. (ii) less 1/rnn 1950 hu11rs (iii) b.:/Wi'<'11 1900 to-
» By data µ 2000 , o 60= = ~~la/> (
x < 45) = 0.31 and P ( ., > 6-t) = 008
·
We have s.n.v : = ~- µ
Wc havc~.n.v , = x - µ x- 2000
.. =-- 0
(i) T f o 60
o ind />( x > 2100 J \Vhen t = 45
If .t = 2100
, z = 100 / 60 .. 1.67 64 - µ
X =: 64, :: = -
336
Thus we have,
p(z <
- - - - - - - -------------..:.:_:~THE:1.fAIJcs.~
z, ) = 0.31
and p ( = > 22 ) = 0.08
ENGINEERING M,<\
~
l]siflg
.
. tJ,egrven
cl>
1sTRIBI.JTIONS
I
:ng we get µ = 48.65 and o = 9.25
=1 = -0.5 and 22 = 1.4 BySoV~•
64
Thus mean = 48.65 and S .D = 9.25
i.e.,
45 - µ == - 0.5 and - µ = 1.4
o cr
69. For tire following normal distribution find c and also the mean and standard deviation of
or µ - 05 o = 45 and µ + 1.4 cr = 64
tire freq11enetJ distribution .
By solving we get µ = 50, cr = 10
- 1 2
Thus mean = 50 and S.D = 10 (X -6T+4)
j(x)=ce24
68. In an ~~mination 7% of st11d1:11ts score less than 35% marks a11d 89% of studntls » We have the p .d .f of the n o rmal distribution
score /es, than_ 60"/o marks. F111d tire 111ea11 and standard deviation if tire rnari.s 111
normally drstnbuted. It is given that if
z .. . (1)
P(·) - _l_
· - fin f e-z212 dz tlre11 P ( 1.2263) = 0.39 and P ( 1.4757) = O.~
0 - 1 2
» Let µ and O be th Consider f(x) = ce24 (x -6x+4)
e mea n a nd S.D of the normal distribution.
By data we have p ( x < 35) = 0.07, p ( x < 60 ) = 0.89 - 1
i.e., f ( x ) = c e24 1 ( x - 3 >2 - 5 I =c 2
e5/ 24 e- ( x - 3 ) / 24
Wehaves.n.v: = x-µ
cr i.e.,
When x = 3S, z = 35- µ
f (x ) =c eS/ 24 e- ( x - 3 ,2/2 ( -Ju ,2 . . . (2)
0 = z1 (say) Cornparin th
0 "' {u pg e_exponents of the exponential function in (1) and (2) we have µ = 3 and
X = 60, z = ~ rov1ded
Ii o = z2 (say)
ence We have
•
l'- r,.;:n: ,... ...... , • '.
I • • • • ..... ,. : :, ,:.,n•i.11 en-r.:l•1l1tv t"IIT't't'
-
tlwt mav bt• tiltt'd t th
- - o r foll .
-~ I ~,TY
~
•. ins arc tos..-.L'<::l . F tnd tlic proba bT
I ity ot· getting (i) c,actlv 3 h
339
-
. • - - ::- l) 10 Olt;~ d..
t 51, co_("") iltka-.t one hcild · eads (11) atlea<,t"\
, ...... • c- - _ 11 :'Iii
•- ·- .. 5 ~ 12 he,,d:, 111 • • •
'r!"f,..·-:........,:. .~
_, c-
('Uf\ e is
- ~
{
The pro •
bability of gernunallon of a •ed in a packet of "4..'l'd
f . .
5. . • c>d arc tal-.cn or e,pcnmentmg on germination in a I bo
Jt 10 :,et!
abilit,· th.it
~ 1s tound to toe O-
' a raton hnd th\'
·
··
l - , , - µ )- : o· prob ' .
:\ \ ) = --= .. (i) seL'<i germinate (ii) atlea!>t !>ero'> germinate
<n:'.:.
w... nt'l"1 to "~mrutt> u and o for the given fn.'quen ,. distribution. X isa binomiallv d1-;tribu~ed rando_m ~ari~ble. lfthemeanand vananc.-ot X are
•\ 1: T .[!_ T 72 + 117 + 8() + 55 + -l 6.
2 and 3/2 resp1..-ct1vdy, find the d1stribution function.
µ=~=-= -- - ==9
_ 10 coins .ire to ed 1024 times and the following lrt.>quem.ie-. are ob't'n ...,l Fit a
':,- ' ' · ~omial di tribution for the data and calculate the C'\f>(-'Ctt'd tr1..-qucncic....
,;- = ~-- - tµ ,2 No. of heads O I 2 3 -4 5 6 7 '> 9 10
3l3o) - 6(-49h9(6-0 +13( 1)+ (100)+5(121)+ -4 (1-4-4) FrequenC) _ 2 10 3 106 _;:.,...
-,- 226 12 59 7 3
~ - (~
er = :!..: ""'13 o = 1.607
s. A swit h bo.ird can hand le only 4 telephone calls per mmute If the mcomm1; .._,,II,
per minute follow a Poisson di tribution with paramet\'r 3, tind the pmbabthh
that the switch board i over ta\ed in an one minute.
:pl = - - - ----- r-l 1-9)~ 5.1666 9. A travel agency has 2 cars which it hire:, daily. The number ot demand, for a car
l 1.o07 ) \2 It on each day i dbtributed as a Poi on variate with mean I :::; Fmd the pt ab1ht,
Thu,, the requu-ed equation of the normal probability cu rvc is that on a particu lnr day (i) there wa:. no demand (ii) a demand 1, reru,
2
/tx) = ( 0.24825 ) e- <.r-9) /5.1666 lO. Fit a Pois on distribution for the following data and caku!Jte the 1 reltcal
fre_g_uencies.
f
- '
0
111
I- 1
63
2
22
3
3
•
_ k_ uitably. e vanance for the following probability distribution~ 11. Find the value of k
uch that th e function
__
.\ _ _ _
0 __ 2 3 I • k \2.,
1 4 I ) / (x) == 1 < .\ < 3
P ( l) k - ~ . lO
~
-----..:.:.._ _ _Sk 10k 10k otherwi e
2. A random Variable X ----- ---- isp a proba b 'I l•ity density function of a continuous random 'Jrt,l
ble · ·c\bo hnd
X == x i has the following probability ma s func~on:..------- ( 1.5 < .\ < 2 ~ )
_ _O_ _1_ _.:_ 2_ _ 3 4 5 6 7 ____,
12. The p.d f f .:, .
· 0 a contmuou random variable..., gi, en b,·
p ( X) _____.-- ·• .
- a 3a 5a 7 5 17•
Deternf -- a 9a 11 a 13a _ 1 ~ P(x)== k .\ (1-x) l , O~x~l
me the value of - -
(i) P ( X a and hence find F· O othen, i-.e
< 3) ("") tnd k
Also f u p(X
ind the distnb .
~ 3 ) (iii) P (0 < X < 5 ) 13. ,.,__ a nd h ence find the mean and the -.tandard d.-, ,att,,n. h·
Ution of x 'ne life of \..nown tot,., ~()0 ml>nt_ ,,
3- Find on an a a compre or manufactured by,, compam •~ i ti1 , pn,t-Jbtlit, tn.1t ::--<'
th bl
e value of k life of av:~age following an e, ponenti~I dbtnbut1on Fin< '
probability ct· . such that ,, ( x) == k 12\ repre¢'
IStribution. r ,= 1 , 2 , 3 ··· mp res or of that compam 1s
(i) le ti md ~:; '<',,r--
!>s than 200 m o nth~ {ii) bet,, ~n 100 nwn ''' - -
~
ENGINEERING MA -rv DISTRIBUTTONS
340 l'HEAttrJcs ~U•• 341
.
14. 1f .1 IS a stan
dard nom,al ,•ariate, find the following probabilities~
Y USUlg no
~. t Probab1·u ty o·1sm·b u ti ons
probability table. .. 'lllal
Join
(i) p ( 0. 7 < x < L (11) p ( - 0.34 < X < 0 .62 )
. ed probability dis tributions associated with a single rando . bl
,ed1SCU d f d m vana e
we ha\rne can
begeneraliz e ortwoormoreran omvariables. We dlSCU
. h d . bl
p b b'li ·
ro a I ty
(iii) P l ., > 0.< :; ) (iv) p ( .i > - 0.65) 1nesa . a ociated wit two ran om vana es referred to 3SJ01nl d,stnout
distributions wns.
15
Th ean w~ght o l 500 students during a medical e amination was f . t probability and Joint probability distribution
· 50el; and S.D weight 6 lg . A urning that the weights are normally;~ lo~ • Jom
find the number of students having weight 1Stributed, U X and y are two discrete random variables, we define the 10ml probab,lrh( function of
(i) between 40 and 50 kg . (ii) more than 60 kg_s. given that
X and Y by
o.,\ 1.67 ) = 0.4525 where <1> ( : ) = &!_ J e- =.~/2 d:: P(X =x, Y = y) =f( :i. , y)
7. l0C ( 1 '2 )1 ( 112 )10 - r = _l_ lO f isd also re fe rred to a s joint probability density fim ctron
- of X and )' ·m the re:,"""·tJ\·e
r ··
2\0 C1 0
X
r er. The set of values of thi function /( " , y ) = / for 1 = 1 '.! 111
'
Expected frequencies : / =}2 I J IJ · To
l , 10 , 45 , 120 , 210, 252 , 210 , 120 , 45 , 10 , I va!u ' ' · · · 11 i called the joint probability d istribution of X and ) . e,e
8. 0.18-17 ~ r e presented in the form of a two way table called the i<1 ml prsihll"lzr:-, tab_!:_
9· 0.2231 , 0.1912
- 06
~ ~1 Y2 -;:--- - ___:~ --
10. p ( x ) = e . ( 0.6 )r
.1 ! - ; l10 , 66, 20 , 4 , l
~ 11 J12 _ __.:'---__ ____.!:_ 11_ - I \ \!)
~
f ,,.2 m! - -
14. 0.~19, 0.3655 Surn -+- -- 1
is. 226 , 24 , o.19n, o.1422 g lY,.)
--- g ( Y1 ) g ( Y2)
~~
•
_ _ _ _ _ _ _ _ _ _ _ _ __
• XEXI'ecta
is a d1scre
·on variance, Covariance and Correlation
. tl te ' ra n dom v ariable
. ta king values x I , x2 ' .. xn hanng
If . J(r) th e n the Exl'cctatw11 of X denoted by E ( X J or µ 15 defined b_
. probab1litv
•· th.e
~
fl • table, f P 1 ), f (.\7),
e •oint p robabililY . - · · · f ( x 111 ) respectivelv
, represents fot1Cllon . X '
I th J f all the entrit'5 in th e f irs t ro w , s eco nd row
JIJu, the
) . ~ ( I/ ) respec11ve1v represents t e s um o f all then tr·· .
um o . h , . II
,( row
S .1/1 . ~ -11~ - ·" ' ~n •es in the~-: µ . == £ ( X) == I: ·\ f ( X;) or I: X f (X )
( ) th I ""' .X i=1
column second .:olumn, .. . 11 co umn.
The Variance of X d e n o ted b y V ( X ) is d cf'ned by the relation
That i;,,
~
2
_t x l = /11 +f12+ · · +1111 ; g(Yi >=lu + / -71+ ··· + / 111 l V ( X) == ( ., ; - µ )2 / P ;) == £ [ ( X - µ ) ]
1
r= I
__ call<.'d
, .. m~r<nnal
- o;-· pr~ b abTty 1 1 d.1stn·b utions of X and Y r~ pectivelv. (Sznu tJ.,."I
µx =E( X ) = L I: ., f ( x , y ) = l: ·\ f ( x, )
are · X 1/ .
, ull« . Mc 71.>m Irk' 11urg111:- of the table the word #111ar8i11nl ~ 1:- /ld11g 11:-i-d) •
= E ( Y) l: y f ( x , !I) == ~ Y, 8 ( Y; l
It shvuld t:'<' nott.'d that
/P1 >- ; t .,2 l + = 1 and
· +f( xm )
I µy
Further E ( X y ) = I:
== I:
,\
x I .If•I
y
JI.I.
I , I
g l Y; l - g l _
v, ) .,. . + g (Y,, ) = 1 :,: • ( X , Y) a nd f (, , y) is the jo int d ist,ibotion ot X ""' L ""
0
ectnt,on of Z in th e joint distrib u tion of X , Y is defined b,· the rel.iuon
nus 1.:, equ1, alent to ,, nting
~ ~ "[ ,: E( Z )= LL . .
©(\ I. , .1/-I ) /··
I I
- - _f p: . !/ ) = '<" ~ ti X and y :i . :ti, eh then the
=: =I • . r ; I Ji J = 1 r .
I /
a re random varia b le:, having mean µX i1fll µ) respe.:
It me.:ins that the total oi all theentnesin
. . the jo int probabi lit\' tab le is ~,qual!C' I· ovanm1cc 0if X and Y den o ted b COV ( X , Y) is defined by the re1auon
Equ1vaJent1
.
. P t1-X
Thatl5P X=x
= i )
= y)
Y- · =P <X-
- .l - ) ·P ( ) . = y) and conversely.
Y w e h ave
F COV(X ' y ) , -_ I: I: X 1/-J-- -µ X µ ) . , ._ -t11• rc>l.itwn
Thi,. • . I , - .ii, J = p ( X =X ) . p ( )' = ) Urther I / I • / I / . ) · ) 1:' delU1t'\ I ,., '
alent to f (
is equ1, , .1/;
~ x,l :
( If; ) = / , 1 m the join t prob a b ility table.
' !he Correl nt 10· 11 of X and Y denoted b)
~
That 1~ to say that ·
to the prod ucc ot it,,. maand. ) a re mciepende
. nt if each entrv / . in the table I> _i#
P (X , Y) = COV(X , ~
rgmal entri e:.. O then, i X and y. .,re I / aid to l,._. , ier""' ax o,
344
-
_ _ _ _ _ _ _ _ _ __ _ _ _ _E_N_G_IN_E_ER
_ I__
NG:._::M:_AT
'..:::!_
HEMAr,~.
""'· ~
345 ,
. dependent ran dom variables then ex)== f J ( x,v )dv
X and y are JJl
Note : If f1 v= - ""
. E(XY ) = E( X) · E(Yl
V ( X Y) = o and hence P ( X , Y) =0
) -- f f (
(i)
II ' y) d ti
(ii) cok . c~V ( X X ) = £ [ ( X - µ X )2 ] = V(X ) = ~ fz (Y
II = - oo
ReJJlar • ' •c; calle d th e marg inal dens ity Junction of x and
V(X) = E [ X -
z
2X µx +µx
l i
f 1 (x ) l-
2 ·s calle d the m arg inal density Junction of y.
= £(x 2)- 2 E ( X ) µx + µx E ( 1 ) Ii (y ) I
. bles x and y are s a id to be in dependent if
V( X) = £ ( x2 ) -2 E ( X) · E ( X ) + [ E ( X) j ·1 -n,e vana
f 1 ( X ) •f 2 ( y ) = f ( X , y )
V( X ) = ~ = E ( X2 ) - [ E(X ) J2 Expectation, variance and Covariance
Nole : Theory and Problems con nected with contin uous ran dom variables is a
iven or the ben t o readers. µx = E ( X) = f f x f ( x, y) dx dy
• Continuous random variables
Let X and Y be two continuous random variables. If f ( x , y) is a real valued
function satisfying the conditions
µy = E ( Y ) = f f y f ( x , y ) dx dy
f ( x, y )'2 0 and f ff( X , y ) dx d y =1
then /( x, ~) is called the joint probabilihJ fu nction or the joint densihJ Junction of the
random vanables X and y .
b d
P ( a $ x $ b' c S y $ d) = J f f (x ' y) d y dx
a C
p(X $ x ) = Fl ( x ) = J f f ( u , v) du d v = E [ ( X - µ x ) ( Y - µ y ) ] = E ( X Y) - µx µ y
is called th . u = - oo V = - oo If X
e marginal distnoution fu nction of X. and y are ind e p endent rando m variables then
If $ ( E ( X y) = E(X
) · E ( Y) and COV ( X , Y) = 0
P ( YSy) = F2(Y) = f yf
f(u,v) dud v ~
distn~~l n 3;Y ~ction of x and y, then the Expectation of <I> ( x, Y) in !he joint
. £ ( y2 ) = ( I 6) ( 3/ 8) + ( 4) ( 3/ ) + ( 49) ( 1/ 4) = 79/ 4
Compult' the_ti1llvwi11s.
I I',, ; . :::
13 _(3)2 = 4 ; c?y = (79/ 4 )-( 1)2 = 75/ 4
( 11 ) E ( X ) nnd E ( Y l (b) E ( X ) ) Hence x
11,us CJx -_ 2 and crY = ✓75/4 = 4.33
(d) COi' ( X , Y) (e) p ( X, Y)
(d) COV ( X , Y) =£(X Y) - µ X µ Y
» The distribution (111nrgi11n/ distrib11tio11) of X a nd Y is a , fo llow .
= ( 3/ 2 ) - ( 3 ) ( 1 ) = - 3/ 2
TI,is distribution is obtained by adding the a ll the resp ective row cntrie and alsolhe COV(X , Y) =- 3/ 2
respective column entries.
~
.\ 1 5
'
I
--,- Y·
j_
-4 2 7 = - 3/ 2 = ---=l._ = - 0.1732
/(.,,) 7 1/2 1/ 2 ( 2) ✓75/4 2 .../75
- I
g ( Y· ) 3/ 8 3/ 8 1/4
I Thus p ( X, Y) =- 0.1732
(a)
E(X) = r. x,f(-';) = (1) (1/2) + (5) (1/2 ) = 3
· blc~.
~2. TI1ejoi11IJ!!..obabilitydistrib11tio11tablcfortworn11dom unnn x and Y 1, nsjollow,:.
_
E ( y) = r. Y g ( Yi )
1 . X ~ _I -2 _ _ _- 1 I _ 4 _ j :,
= ( - 4) ( 3/ 8) + ( 2) ( 3/ 8) + ( 7) ( 1/4) =1 ~, I 0.1 - -0.2- - - r - 0 - - 0.3
Thus
µx = E ( X ) = 3 and µy = E ( y) = 1 ~2 . 0.2 0.1 - , 0.1_ _ _ 0
= (l ) (-
4
) ( 118 ) + ( J ) ( 2 ) (1/ 4 ) + ( J ) ( 7 ) ( 1/ 8 ) (n) E:rpectntions of X, y and x y (b) S.D.s of X, Y
(c) CoV · d Y
5 4 nrinnce of X and Y (d) Correlation of X nn ·
+ ( ) ( - ) (1/ 4 ) + ( 5) ( 2) (118) + ( 5) ( 7) ( 1/8 ) 11
F rther .~ · bl
Alsoft Ver&y that X and Y are dependent rnndo111 vnr,n es.
1 1 7 lld p ( X + y > 0)
=- - + - 5 35 3
2 2+s-s +4+ s =2 :>:, lvt . . II the respective row
Thus entrie argina I dis tributio n s o f X a nd Y are got by adding a
nd
C( X Y) = 312 s a th e respe c tive co lum n entrie .
~ L I T Y DISTRIBUTIONS
..,
.'I, - 2 ~
I 1tion of X ,md ) p(X,l)- COV1 X, Yi
O.o tlA S l _v,) 0.3 0.3 (dl Corf\' •
I ( ', ) 0.1
OJ
(a) µ, = E \ .\ l = I 1 I l ', ) , ))
-o, =-(l1
I( p( X , - < o.4q H 3.1 l
= \ I l l ll.o l + ( 1 l ( OA l = 1A
p( , Y) = - 0.3
~ll = [ ( ) ) = - .1/1 S ( .1/1 )
U X .ind y arc indcpl•ndl'nl random vanablt."> we must have I t , , '{ 1 11 1 = /
= \ - _ ) ( 0.3 ) + l - 1 ) ( 0.3 ) + ( 4 ) ( 0.1 ) .,. ( 5 ) ( 0.3)
= - 0.o - 0.3 + 0.-1 + 1.5 = I llranbesccn lhal f( 1
1 )g(y 1 ) - (0.6)(0.3) = 0 1, and /
11
=01
E \ X) l = I t, v, /11
I(
= l 1 ) ( - 2 l ( 0. 1 l + ( 1 ) ( - 1 ) ( 0 .2 l +- ( 1 ) ( 4) ( Ol
Sumlarly for olhl.'r. ,11-,o the condtllon 1!. nol sattsfieJ .
.,. ( 1 ) ( S l ( 0.3 l .,. ( 2 ) ( - 2 ) I O 2 l + ( 2 ) ( - I J ( OI
+ ( 2 l ( 4 ) ( 0. 1 l + ( 2 ) ( S ) I O l llmre we conclude Ih a t X and )' are dependent random variables.
= -0.2-0. 2+0+ 1.5 -0. -0.2+0.8+0 0 .lJ
Thus E ( X l = 1.4 , E ( }') = 1 and E ( Y) = 0.9
\\'dlil\l' X = ! ',
(b)
al = E ( X2 ) - µ~ and ~ = f ( )2) _ µ;
E l X2 l = I \ 2 fl \ l Also
I I
/ 11 = 0.1 , In= 0.2 , In= o, /14 = 0.3
lh,· ~•"'" data and the r<.' 9u1red /, 1 1s exhibited in the fo llowing table. = 4.55
.. , I/) 0.3 0.5 0.2 74. X and Y are 111d,•l'c11de11t random t'tlnab/._,... X taJ.,: t•a/m, 2, 'i, 7 u:1th rr •rcbi:::,
I 2, 1/ 4 , 1/4 re,peclm.'ly Y taJ..,· ,•aliu< 3. -1. :i with the prub,,b:1·::.i
Ill= (0.7)(0.3) 0.21, l,2 = (0.7)(05) = 0.35, 1/ 3, 1/3 , 1/ 3 .
In - 1 0,7 Jc o 2 J = 0.14 (a) F111d the 10111t pro/,a/11/1l11 d1stnb11t1,m ,,f X ,md ) .
111 - < 0 1 J ( 0.3 J -= 0.()9 , 121 = ( 0.3) ( 0.5 J = 0.15, lb) Sliou, that tire COl'ann11ce of X a11d ) 1< <'1711111 t,, :a11.
.'<I l/ J
l' h.1\ l' '
,1 - I( ., ,)· :,: ( y ) whcrl' i , J• = I , 2 , 3
03 1
Cl.'> 0.2
1,, 6, Ir~ = 116, / r3 = 116
l1, = 12 , /,, = I 12 , 121 = 1/ 12
X 4
5 --:---._ Q
-- --
' ( .t 3
l
1, 6 L6
6 -- y
1 24
1 I:!
:,.
l '12 l 12 I 8
-
l/U 0
I{ 2 I 4
--
1/12 I 4
l / 12 1' 12
- -----
14
--
-l 1 :!-l
I 12
1 3
----- 1 S
--
3 I 3 0
I
» ~taf%inal d i,tnbullOlb ol X and ) i,, .,~ folio" "·
3
0
6
y,
.... i~ = E 2 I 3 It>
\ I 2
I
1 4
g ( Y, l
= ' 1 -l
- 1'2 ) - ( ''>I
~ 11 4 ) (-
~ =c. )1= 'I"
-:- 11, g ( y ) - l ) ( l • ) = -t CO\' ( X, Y) = E < X Y)- µx µ,
= 3I l , )
£ ( X) = L t , f (:l, )
- X ) , _- ~ .:>X -1' (/ -1 )( 13 ) - ( :,
- ) ( 1/3 ) = 4 I
= ( 2) ( 1/-1 ) .._ ( -l ) ( 1/2) + ( 6 )l I -l ) = -l
! I
-- '.,... , -
- I I 3 I ( 1/ 6 )
- (2)( 4 ) ( I/
£())= -YI O<> (1/)
I
• I
__J
We note that r PIZJ = J 1/6 - V U ---V 912 - 10
1/6
,- 354 ENGINEERING
MA n11.:1.y. IJAB!UTY DISTRIBUTIONS
f-(-
x , y) = I: ( 2x ~ If) and the join t probability distribution t bl . = ( 0) + ( 1 ) ( 9/42 ) + ( 4 )( 12/42) + ( 9) ( 15/ 42 ) = 192/42 == 32/7
- - - __ a e1sforrned 2
X
Y
O 1 2
- as~
3 -----.:.::.: (cl c?: = E ( x2 ) _ 2 = 17 _ ( 29 ) == 2.30 . 0 _ ✓230
- - - - ~ - - - - - - - -__:::__ _j__ Surn x µx 7 21 441 ' x- 21
0 0 k 2k
3k ---- 2
2k
- - - - 4k _ _ _ 5k ---
3k ~14k
cl: = E ( y2 ) _
Y µY
2 = 32 _ ( 39 )
7 21
== 495 .
441 ' cry -
_ \1495
21
2 4.k 5k 6k
7k Thus crx = 0.72 and cry = 1.06
um 6k _ _ _ 9k 12k
15k
(a) \\e must have 42 k = 1 . 7s. Consider the joint probability distribution as in Example - 6 and find
.. k == 1142
(bJ ~1arg1nal probability distribution is as follows. (a) P( X = 1, Y = 2) (b) P ( X == 2 , Y = 1 )
(c)
x, 0 1 P(X~ l , YS2) (d) P(X+Y > 2)
2 0 2 3
J<x;J 6/ 42 14/~ 42 » Let X = { xi } = {x 1 , x , x } = l O, 1 , 2 respectively.
2 3
( = 117 =1/ 3 =11121 g (Yi ) 6/ 42 9/42 12/42 15/42
y = {Yj } = {Y1 , y 2 , y3 , y4 } = 1 0 , 1, 2 , 3 respectively.
cJ It can bee351·1
Yseen that J(
- 1/7 I _:3
- ~
;,/ 14 =2/7 1 =:,"/14
Hence the rand . xi) g ( Yj ) ~ Ji . Also J11 =0, 112 = 1/ 42 , /13 == 2/42 , J, • = 31-12
- - - om vanables are dependent. /
121 = 2/42' 122 = 3/ 42 , /23 = 4/42 , !24 = 51-12
n. Consider the . . - - - - - - - -- -- - - - - - -
Joint probability distrib . . .
(a) E ( x) E ( y ution as in Example - 6 and compute theJol/()1lllng. / 31 = 4/42 , !32 == 5/ 42 , /33 = 6/ -12 , ]34 = 7142
» (a) £ ( X ~ ) , E ( X y) (b) E ( x2), E ( y2) ( ) (a)
P(X == 1 , y =2) ==/(x , y ) = 123 = 4/ 42 = 2/21
J- r xj( C <JX, <JY 2 3
I Xi) (b)
1
p ( X == 2, y =l ) = f ( X3 ' Yi) = /32 = 5/ 42
- (0 (c)
E(Y): )(6/42)+(l
- r Y1 g(y. ) ) ( 14142 )+ ( 2) ( 22/42) = 58/ 42 == 29121 P(X~l , YS2)
J I
- (0 ( X' Y) take the value
E(Xy~ )(6/42)+(l
1317 l'h (1, 0) (1, 1) ( 1, 2) , (2, 0) (2 , 1 ) ( 2 , 2)
= t -\Y· /. . )( 9142 )+(2)( 12/ 42)+(3)(15/ 42)=
j ) j
I . I
ey are respec t·1ve Iy
::: ( 0) ) {\ 1/, ) ( X , , }/3 )
( x 2 , Y1 ) ( x2, !12) (
_ + ( O+ 3142 + ~zj~Zl Xz, Y3) ( X3, !11 . J, . - .,
- 102142 _ 8142 + 15/ 42) + ( O + 10/ 42 + 24/42 +
- 1717
•
356 ENGINEERING MATHEM/47.,...
~
p ( X ~ 1 , Y S 2 ) = 121 + 121 + 123 + /31 + /32 + /33
2 3 4 4 5 6 24 4
= -+-+-+-+-+ - = - - -
42 42 42 42 42 42 42 - 7
Thus P ( X +Y > 2 ) = 5/7 / = P(X = 0 , y = 2) = 2/8 = 1/ 4 ; out comes are HHT and HTH
13
79. A fair coin is tossed thrice. TTre random variables X a nd Y arc defined as follows.
/ =P(X =
14
0, Y = 3) = 1/ 8; outcome is HHH
X = 0 or 1 according as /read or tail occurs on tire first to~s. /21 = P ( X = 1, Y = O) = 1/8 ; outcome is TT T
Y = Number of/reads. In= P(X = 1, y = 1) = 2/8 = 1/4 ; outcomes are THT , TTH
(a) Detenrrine the distributions of X and y
(b) Determine the joint distribution of x and y 123 = P ( X = 1 , Y = 2 ) = 1 / 8 ; o utcome is TH H
(c) Obtain the expectations of X, Y and X Y. Also find S.Ds of X and Y 121 = P ( X = l, y = 3) = o ; since the outcome is impassible.
(d) Compute covariance and correlation of X and y
(Thise values can be wrrtte11
. ·
quickly by look.111g at tire table oif 5, X' Y)
: J:ef samp_le space S and the association of random variables X and Y is gii•en
.!.._ e ollowmg table.
---
The req · . . . X d y is as follo\\S. --
U1red Jomt probability distribuhon of an -
Sum
__s__ TTH
-
H_
HH
X
~__:_:_
_
HHT
o
HTH
-:::-- r -o·___,_.::_
HIT
o _;_~o_j_~l:__J__.!_
THH
1 _~~-t
THT
--:.-...:.....:...-1-_:_:.::..:.-1-_.:..:.~+--1 0
T V8
2
V4
3 -
1/8
--
J/2
y 3 - 1/ 2
'-- _,_ 2 _J___ 2 1 2 1 vs O ----
(a) The probability d" • . '_ _L__:: _ _L._ _ VS V4 - - -
st .....___ Su_rn_ _-L_ __., ---1--- 3/ 8 118- - -I
' nbutron of X and Y is found as follows.
X= j x ' - 10 1 kl VS 3/ 8 --
; - ' and Y = / = I y.}I
1 2 3) o' µ - - -,-
P( X = 0 ) is 418
I I
x - E(X) = :E x.I /(l I-)
p( y _ O . = 112 ; P ( X = 1 ) is 4/8 = 1/ 2 i
"' (0)( 1/ 2) + (1 )( 1/ 2) == V 2
p(y =~~2
118
• P ( Y = 1 ) is 3/8,
µy = E ( y) = :E Y, g ( Y; )
- IS 3/8, p ( y = 3 ) is 1/ 8
:~ :_____________________ E_N_G_II_IE_:E__R.::_11~
1"- 1 ,.
3_ ' '" nu ,,~, 359
" l • 1 l1 1 l ., 1- l2)111 )+ ( 1JC I XJ : 12, <,
~ Pf \ t I , y ~ 2 I = 121 ... / 22 --- /2.,
3/ 2
r \ ' !. 1 I/ 'l -1 1-I-I M= I'.!
Pf X ~ I , y ~ 2) = 1/ 2
• Ill • I I ~ 2 •Il l 1/2
\: I µ; ,ind r? I I ) :. l
p 1X:5 1, ) :5 2J
l , tJl..t' the \ alu e-.
· I ~I I -I O IJI0 , 2 Jl l,U)( l , l )(l , 2 )
2 wd o 1 ., 2
PIX :5 1 , Y :5 2 ) = /11 +/12 - /13 - /: - J:.:. - f:.1
ldl I 1.•. , l µ, µ :0 ..-1/ ..- 1-1 ---18 +1-l - l =71
II
! : l ~ I • fl-.i, P( X :5 1 , Y :5 2 ) = 7/ 8
) PIX.,.)~ 2 1
\_ di
) ) take the\ alue-. I U , 2 1 I O, 3 J ( I , I 1 I I 2 l I I >l
4 I :1lt'I are r~pt.'Ctl\ el~
.; '1 ( 11' 11, ) ( \ 1' 1/4 ) I '2 • .1/2) ( \2 ' Y,) ( 1.~ ' 11
4)
((JI \ , l - I 4 1nd p I X , - I ,3 PI X - Y ;?: 2 ) / , .,. / 14 ..- /22 .,..•/~, - /,4
' >. aud 1 a, 111 I \a1111•f,• ~,, (, 1111 ttl( t~ i ,I = 1/ -l ..-l/ h+l -1 ..-1 -. . l]:3 -1
_ l' I X-) ;?: 2 ) 3/ -1
l u/ 11,, I
•. , 2 I
,,, - IJ J ~·~ ~ I 2I
'I . \
• dOt
I )
a,,. 111d,·1•,·11dt·11I rn11d,•111 .,1r1,:r ,....
I
I'
., :it,· ' •11' a·111·, ,..,u/1,.
f 1
11.\ 1 I C)
I/ I • l ( ,\) I
'"' l (XJ 1 ( )
360
\ µ , r ( X} ) • -~
= E( X ) £ ( )
l-E(X
- E( X l £ ~
(Y)
'tt-t
--.....;:
problelflS on conti nuous random variables
n=o >E( } · >==o 82- x and Y ar,· ra11d,,111 t•ariablc:i liat.•inx ;01111 densit11 junct1cm
Thu~ COl"t X .
-1.\_1/ , !l ~ l S I • ll S _II :5 I
(c) er..\ - i -- -.. : txI -,. .11; r' J I - µ X2 ~ } f l-' ' Y) == { 0 , ,,t /1c•rw1,;,•
' , I
,
q
_v I
ll
=I
' -
I X-
r
,
• I ) -- :!: s<.11, > ~ ,
.t; I ( .l , l >>
E( X ) = J J x f ( .\ , I/ ) d:t d11 == f
But
~
~
1
• , ,- f ( .\ )
r; I, J = '
mularl\ - ,
- I v;l ' 1 = -" >
• 1
I i.t., = J-4 [ -~ ] I I/ d I/ :: -4 y_2 1 = 2 r
Al'<> ! 1
11·
•1 ~ <_v1 )
• y = ll , =0 · . 3 L '.! 3
I
.,
I
= -l" r f_v, J. ,
J fy
.I. -t , 11 d.t dv
. = 3
-
fun"--
U~f
. . I • ( t· )
µ X - } =£( X• I
-,.} )= !
_.-
.,
I/ g(y,>= E tX )
£( } )
[ ( )' ) = J J y/( .l , y ) d., ,ty =
I I
( .\ I + .V, ) / i J
L'sinoa11 th ( X - Y ) =£ 1x
~ "
r._,.,,ultsinth )+£ ( >')
·l "' ! .r f <.I. , ... .,£ eRfiS of (1),
. Weh
• (X avt,
' >£<Yi ,
0-:- + >
.\-i "'E <xi >-2r ,x - .V ,-x<v)
I ., - E( X)+E (Y):? 3
- (E ) £CY )
<r. 1X ! J _ .,. E ( }' 2)
.\' ~ y
l.e. ,
Thu~ ~
=
£r xi >- [
{ t ( }·) 12
2£I X )
E( y) = f f ., y f ( t , I/ l d., dy
£<n = J Jy J( x, 11) dx dy
I (cl _ c,o - 00
-t
Jy XJ/ dx dy
lt (an ~ Sl'<'n that, 96
E ( X + Y) = E ( X ) + E ( Y ) = 4/3 y = l\ = 0
E ( X Y ) = E ( X ) · E ( Y ) = 4/9
= 1-
96
f [? l 2
dy
83. Tlrl' jo111I dc11~1ty .f1111ctu111
Fi11d
f (·' ' Y) - O l
(11) tltc Pal m• (1( k
()/ /rcnt•isc
(f IW(l etmli1111m1~ ra11do111 vn rinh/c~
_ J kx y ; 0~ ., ~ 4 , 1<y<5
(b) E ( X) (c)
X a nd Y is given i,11
E ( Y)
= -
1
96
y= I
8
[ 1/
~
3
I -
I
=
I
36
_ 12• 31
(125 - 1 )=36 =9
I
y = l ., = O
s r y-,
i.e., J l- y d11 = 1
k
y = I 2
I
= - 5
96
J [ -33 Y 2
I/ dy
·
fi f y=I
1
-:6 ~ [·n
i.t',,
I.. L·2 = 1 or • k ( 25 - 1 ) = 1 k = 1/96 2 2-t
( 125-1) = ? 7
Thu~ U1e required k = 1/96 = 27 -
(b)
E(X) = J J x f (.,, y) dx dy
'!nus E ( X y ) = 248/27
-00 - 00
5 .;
(e)
E(2X+3Y) = J J (2, + 3y)J(r , y ) d., rly
= J I ;\ \ 1/
96
dt dy
y=l x =O
5 -t
= J J (2 , + 3!/) -~ dx dy
v = 1., =o .i 1
Thus E ( X ) =
=..!...
96
sit=
5
3 y dy = 2[L2I = 24
J64
I
9 2 1 9
=8
3
= ~ .. I,{[2; i: 't, 1
y +[ j \ .,y
364 ENGINEERING MA
THe~T/cs
;; 5 ~
[~ l
-
I
+ 24 [ t3 J ] ::: 1~
1
9(,
= :!I
3
I
Thus f(2X + 3l' ) = -li i 3
x2 y il3
1/2 [
= 2
3 1/2 [
f
x-., 7
2 + 24
l dx
8-l. UX .m~i ) 11rr t'\111ti111ww: rami,m1 m ria!,lt-:,: having thr 1v111t dt'11sity Jimctiou l =0
') ')
. c ( .\- + v- ) ; U :5 x S 1 , ll S I/ S I
t \\ , 11)= . .
· 0, vt/11·,wi::t·
,frtrm1i11e (i) c,,,ista11t c (ii) P ( X < 1/2 , Y > 1/ 2 )
(iii) P ( 1/..\ < X < 314 ) ( fr) P ( ) . < 1/2 l Thus P(X < 1/2 , \' > 1/ 2 ) = 1/ 4
= J f ~2 (x2 + / ) dy dx
1 X = 114 y- 0
1.e J J re ., 2 + i) dy d., =
t=Oy=O ,? 11
~ x = Jl/4 x2 I/+ 3
3/ 4 [
- = o dx
l r 3
I
i.e.. C f
• l -'2 .1/+ !!..3 d., = 1
l =0 = 0
I
., l I
~
z.e c J I X- -
3
d:t = I 3 1
t =OL J
:::~ f13 . . ~ 7= 29
2 L96 6J 0•
lhus
Thus c = 3/ 2 P ( l / 4 < X < 3/ 4 ) = 29/ 6-l
rY DISTRIBUTIONS
366 ENGINEERING MATH ROEJABILI
_ , ~ ~ - - - - - - - - - - - -- - -------..:.:. EMAr1c.s pflT p
- ~
(iv ) P(Y<l/2) 0, X ~ 0
I 1/ 2
= f f ¾(x2 +y2) dydx F (X) ::: ~(x2+1), 0 Sxs1
2
x=Oy=O 1
1 , X?. 1 7
j
2
(b) Since f ( X ' Y ) -- i2 (x + / ) is a symmetric function and o <-" y <- 1 we have,
0I y $ 0
1
F ( y) = P ( Y S y)
2
= 2 (if+ y), 0 $ y $ 1
1 y ?. 1
J J f ( x, y ) dx dy = l
U =-oo V=-oo
X 1 2 3
i.e., J J ~ (u +v2) dv du
2
i.e., J J c( 2x + y ) dy dx =l
u=O v= O X=O Y=O
i.e.,
cf [ 2xy+
X=O
i_I
2 =0
dx = 1
2
i.e..
c
I( 9)
6x+ - dx = I
X=O 2
thus the
'[,x2 + ~ I a ] c ( 12 + 9) ::: 1 O
r C : 1/21
~ "quU-ed C a 1/ 21
DISTRIBUTIONS
~~----------
368 ---- ----.:..:.:~ ENGINEERING MA THE
ptif_~BILI
,;2
r-x
p ( l 2'. J , _v $ 2) =
2
. -,."
J .
1
ll ( 2r + .1/ ) dy ,fr
== _J_ f
2
x
[
2 dx
:-.ow. 96 X =0 = l
t = ly= O
2 2
== 1 f x [ (3-x>2- 1] dx = J~ f (8.r - 6x2+r3) dx i
192x = 0 O ,5
_
1
21
X
2
J (4x+2)dx
= I
_1_
== 192 [
x
if-2x3+-
4
4
I = _ 4 = _]
192 48
6
5
4
y
5+ - - - - - - l - -
y=5
I"· f f 2
X=O y =O
X
c x y dy dx =1
y= I
I,. '.!:[fL. dx=l
2
X i.e.,
~ f x3 dx = 1.
0
P ( X + y < 3) = f f f ( x , y) dx dy X:
R
2 3- x
= I I !.1
96 dy dx
X=O y = l
370. ENGINEERING MATHEM
:.:...=------------------------.::::ATics.lv
1 X
=½ Jx [fr·dx = ¾ f x
.\ = 1/2 X
3
= 1/ 2
dx
15
.5
1
= 16 [ x4 i~ = 1 1
16 [ l - 16 ] = ;;6
I 3- T
f J i (6 -x - y) dy dr
Thus P ( 1/2 < X < 1 ) = 15/ 256
p ( X + Y < 3) =
l =0 .V=2
89. If the joint probability function of t/ze random variables X and Y is given by
r-
f(x, y) =
< 1, Y < 3 )
0 < X
othenvise
(b) P ( X + Y < 3)
< 2, 2 < y < 4 =
=½f
½f
r=0
x=0
[ 6 !f - .\ !f - I~
[6 ( 1 -x)-x( l- x )-~(3-d+ z] dx
x
=2
dx
1 3 - 16 f ( ,, 2-8x+7) d:1
J J ½(6 - x - y ) dy dx \=0
=
1 = O y =2
= ..!_ [ x3
16 3
- 4..,2 + 7:r. l~ 1
=
24
~X+Y<3) 5/ 24 -
random v~riable,
90 . 11 0" trco co11t11111011,
· Suppose t/zat t/ze joint density f1111ctro 1
X and y is
... - '
:.
Pl l < :\ 1 - j . j,,
htr
:r • I
~ ti( ,h
\ w• -
b
•• o
• -,I
;
= ~ -
3
- :. 1
:4 :4 ~ - j \
"\
. 1·
~
.h = ;
1
::i .,. .:!4
0
. n=s
= •• -
:-1 - -• P( Y < X l = i --'
E.XEROSES
- ·:! ; 5
t Jyndom , an.1H,~ X .1nJ ) ha, e the fol.lo--'lllg JOll'l dJ5tri,cDCn
1
lo 12 4 6 -3
-c X
iJ.1
0I
0.1
0.3
t
Fr.l (a) mal)!1rul Ji-tnburion!, of X anJ )
/bl cov1 :\ , , 1,i corttktionol x .inJ ,
M the 'an.1bl,-. X y i.nd~ent nnt1om ' ~ '
2. Let x_b.,aranJom ,anabl,•"1ththl'follo"~Ji;;tfibutolar,J l
be x·
X .:! I t
I
I .
!< x, ) 4