0 ratings0% found this document useful (0 votes) 67 views16 pagesQB Unit-5
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
UNIT-5 CORRELATION AND REGRESSION
4 | Givaxk)
Q)| unite the formula for Kart Pearsons cockgicient correlation »
There axe sevenal formulas to calewake Kan) peovsons coet ficient
of comelakion ()
OE Coe NOISE Sea
wan
TH Ty
Ae
NAG
Here x=%-K, Yay-V
whee %)F are means oh series X49 Y
Tq > Sd sles %
Ty > 5D seuies Ye
4B} Define Regression.
The statistical ymethad ushich helps us bo estimake, the unknown
value of ons variable from the kroton, value sh the culated variable
1s cotled Regression”
Ag] waite the formula for yank corselation (Speonmans Tonk comseladion)
BG 2D , ;
We Nee) foronk coehicient of correlakion
Bosum ob sqpanes ¢ ditherencs & too
ranks
N—> Wumben of patied obs envations.
f= eee say what 4
Rar : :
Pomerat tert recieeriat crt an
Ady
waite the applications & Regression.
Xb v used to estimake the melation bekooe two economic
variates Like Income and expenditare-
Tew Highly valucble tool iy economics and Cusiness,
Te B, videly used for prediction punpose .
¥ we can calewlake coefficient ef cormelation and coetfiefenk to
dekerminadion uitth the ‘help oh the negresion coekhi clertbsAe)
2a)
following dake,
Write the formula for the vegnesion eqpation of Xon Y.
f 2% = Nat bzy )
| ,
| 2x =aZVG bzy> |
ey solving ue get [X= arby|
Ap the actual mean & fraction actual mean uy not tvaction:
KR aT. TRG xK 1H (Vy
vy (v-¥) yO)
co Ss pre 4, SK. EK
vy Se ree vy zy |
Eady? — (Zdyy |
een moareren |
(3manks
Find *he coefficient of comvelation bebs
wea KX and y for the
xX w /WD \s | 26 123 | oF
y [V3] 18 Jia] 25 30 | 10
Aek, c= % Values
B=Y values
The actual means ane
X= HG
n Seite.
Pana CP ae
“Ty oy
oa y | xex-® | y-y-v
<5
°
-6
+
12
82)
The cootficiant of corselation is
ee ZK _~ 40
os
Jeezy — Yoaexste
eb 6 p5,
243-04 4
Fromm the sampls of 200 GeortO pains eetesenmend Following
Qwonkties we calatated.
EX= Wesy, Lys 20-42, Zx*= 12-16, Fy% [Link] , Z¥Y¥= 22-13
Frm the above daka show hou to compuke che coefficient of the.
eqpiations y=o4bx
dot,
the required skvaight Jine
Y= atbx
The two normal eqs ave SY=bzx4 Na
i <8XY = bax 402 ~
Substituting values,
20-42 = U- aub+ Pao
22NZ = \216b+ IV 34a —>@
Fro eyD
20/42 ~\W34b
O= ®
200
su @®in ® |
> 2213 = 12 16h W342 sub)
> 224 —
=) AS = (21Gb 4 11.34) (0.1036 — 0. SG4 b)
=> 22.13 = Vl45S—0,643b 4 12-16b
=> 20.955 = \1-5\4b
> b= 20.955
te SUF
= 20-42-34 (1-829
eee OO pucaeacl
= 01036 ~ 0.0564 (1-82)
> R=wE]20)
The eqpration of straight dine
Y= 0.00054 1-82.
Calculate the wegression equations a Yon x
below taking dovrations from actual means
from the doko given
% Xony,
Price jolla
to | 22.
3
uy
Amount dormandect
Steed
XY = amount damanded
The actual means aie
2 | 46 [45
2 _| 29 | 33
ZH
N
YM |axnx-® [ays yy | dx? Jaye | dody
10 3 a1 4 |ia6 |) 42
22 BI 2 \ oy a
24 ° ° ° °
24 = 3 oa
24 +3 5 | >
aS 2 q 4 | 81 Ls
cee ee SS
Regression eM OL Yon x b
Y-¥ = byx (X-3)
Sha
byx = Zdordy — zy ae oxy
gd? (eda ay- 0
i
AY ~ 208420)
Regresion equadion &
3-084 (X-13)
= V=2N + B-OB4Y — 40-092
Regression eqn of x on YB
x-X = bay (Y-¥)
wy = ZY = Edn aay (yin h(a) :
you =
© 3dy> (sayy
N
Reqnenion equation is
X~\3 = 0-235 (V-24)
=) X=13 40-2259 -5-64
Bifeance bekween correlation and Regression,
Comelakion
Regression
% Measures the strength and
divection a a Linear swlation-
-ship Eetwean two vordables.
# The relationship UW symmetric
Abe comelation ib same ushether
you treat ere vaslobl as
\ndeperdank / dependent variable.
% Rests In comratiort conident
Ge) bekwan - and +1, Indicads
strength @& diveckion of the
aclatienshi p,
# Predicts the value of one
vortable based eh another
Variable
K The sulationstyp & assymetrye
ont vattabl & censidened the
\doperdent and the othen is
dependenk+
X Provder an eqpation $yemuc
thak dasceibes the sulationship,
allowing gor prudichers ef
dependant voriabl. bored on
Independent vorioble,2.0)
he rank o& 16 students in Mathernakics and Statistics ane as follows
Cty s(2yt0d, ,39, C4, HD 10S)59 5 (6,99, (4129) (8,6), (4,89, Co, 1535
(12,0 1 03,14, C4129, C1S)16), (16, 139 - Cocwrade he wank correlation
coufticient for erdticiencies ob ths group In Sathernakics ard the
Stockists
jek, x-nank of mathematics
Yq sank of Stokistics -
pe
°
64
°
°
°
1
25
e |e a!
ae | u
10 nn Sal \ ;
q is “4 16
2 S 3 Gh
2 | 4 a |
14 (2 2 &)
is] te -l 1
one cl
136
The wank rselotion 4s
Je \-\eD7
Nur)30)
208
o BU) _ BHI
| rte DF es
€
= 354 _ Bog
econ
2. The wank covvelation & 0-8
l=
moxks’
Colewlate the coefficient of correlation beturen ott cars and
annual tairkenanice dest and commen + ”
aE 4 [6 ) els | o 1D
Anmuiol cast | 1600 | 1500 | 1800 | 1400 1400] 2100| 2000
ME, op age hy cons
Y= Annual cost:
the ackual means artThe Korl person's coelficient ef covvelation &
op EY aE:
Vex? zy Yr0x2
= 34
= ——— =0.-83!
qua Sy
3B)| ind Kon peonson's coefficient of correlation from Ue
Following data-
Woges 100 | tot | to2 | 192] too] aa] ax | ae | ae | as
Cost Vivirg | as | 29 | aa | a4 [as | a2
a5 | a4 | 40] 4
dak) y= wooges
y= cos of Living
The actual means arc
K= ZH _ 990
=4
N 10 A
y= 2M 2 450 -45
N&
Ale Kou pearsonis coriticient of correlation i
mea) ect
peo enon
= amma
Deteuviine the equodion of a straight Une which bet fits the dato
x [wl oe] Le [al pos
y | to | 22) 24 | 24 2a | 33 34
du,
> the neapised straight Une is
Y= a4 bx
he two norral equations are Fy = bzX 4 No
ERY sbZx? 4 zx
x2
7
| 100
vay
169,
2s6
284
4oo
625
Ls ver | aise 1983
Substituting the values
sy = bgXANO
ZY =182, ZX=M3, NaF
Nsbe-+ Fa =182—>O
ZXY = 3186, ZX*=1992, TX=NB
19g3b + Za = 3186 —>@3d)
ey K 13
127696 + F410 = 20666 le}
eey® X +
138816 + 4a\a = 22302 —>@)
%O-@
=> pe 1386 _
wi
>a=082 _ .
fan eh steaight tine & y-a4 bX
1-56
Q=0 82"
beuse
0-824 LSéh
Find the most Ukely production a aaeat toa rainfall ao
from the follsing dato
Rainfall | Production
| Average 20 SEs
Stardasd devotion too kgs.
corfichent & covrelotion| 0/3
Weave to calculate the value gf y
when X= 40
So, we chowd Find sragteriion equation & ¥ on x
mean ef X- seties x) = 30
wean e& Y- Seties Cy) = BOO.
S-D A X-sortes = 5
SD A Y-series Ty = 100
Regression ob Y on X
Y= BSR (xX
Sh Fh (ex)
Y-Sd0 = 0-8 (&) (x-309
X= U0,
Y-s00 = = (40-30)V-S0o7es
von 1)
. ee
= y-500 = 4
=| Y=s00+0-4
+, The expected value Oh Y ‘& S00- Ukg-
32) Following are the sank obtained by to studers in two subject,
Stokistics (x) and Mathematics LY). To whak extenk the knovoledge
of the Student in two subjeds bs sielaked,
SE a ee es
| veeR Tie ish] 2) lab ea ioe semis
| 4 seonank & staintics
| Y= nonk oh mathematics
| [ex [iy | oes
| roman ez ae 7
| 3 \ 2
cf S S|
| ran eons ee :
6 q -3
4 +
& to oe
ot 6
10 g 2
7
“Whe wank conetation
ozp>
en
AUD
&4a)
mace
| 1D
Sohne
Vrs
(0 ranks)
Eetiowings Colculode the Kort Peancons Coefficient ef correlation for
| the following poined dake -Whak nferedce would’ you drat foiom
Nreno. oh rems =(0
| the enna) as
x | 28] 81] 9] 38 | a5-| 23] 40] 32] 36 | ss
L y_ | 23 [34] a3 [as [ 30 |e | 28 | a [36 | 38]
ON > ZAI BSC ace 36
N \o
Y= Zui = 313 2 3a, 8 3 . - ‘
N 1 i
x Yo xan | ye y-¥ x2 | y?
2S a 8 ey | 64
4) | 34 s 3 en ey
4o 33 q lo
ae | 34 2 3 ay 4
as | 30 Fi bm 1 1
33 26 oo Ss 4 | 26
40 28 4 = 16 a
2am a ° 6 | ©
36 | 36 ° s Omen |S
33 | 38 3 7 a | 4a
356 33 -4 3 1eo | I4sAa 1 9.440
VAS +23
Yb)! A sample fh \2 haxners Le) and thein elden cons gave the
folowing dado abeut thei eldan sons (S)-colculote the coefftenk
& Comelakion » 2
€ [| es [3 ] ot ]ea] ce
s | 68 [ce | cs | os [ea
positions | | and | a] ot | st
Ea Fe1| zo | 64 st
wmnks \ ao = as
positions | Se] ott} ard) wt | sth] ob | ath] sth] ah roth) qth ah
eat M |] ao | 61] 68 {es | ce }os| cartel col es | es
sans 1 2 |B Iss|sstss [Ss] se lashes] ws | us| Fathessix) | somiyy | x | y | Dewy
6s 6s q4 |SS} 3-5
63 66 we 4d eS
G4 68 6S ISS} LO
GY 6s 1 [US} --s
cs Gl) 1 Gisife || ws
2 66 12 las} os
ie 68 | 2 |ss\-as
oS 6s BQ jws) as
& a fast, | sc
ce ot |@s|e | ts
cy 68 | BE} o5] -2.¢
a 70 aes
To X-seties fox 68, m2.
fow 64, maz
2
Y-se’es for &, may ” :
fos Ge, m= 2.
fos 6S,m=2
The covvelation t (2) 4 (2%2)
v VD
The correlation x Wha) 22) , 242)
io ipa 2
eee Cae
ie Dan la ea
The nank cometation coefichont is
= 1- Gf 2% mim , Wi
ian [ sso ETO
wens
1- ©G2-S4146) =~ 645)
wa ae
286
640)| civen Bwattote dato
ifs ]3fa)y [fale
ly fe [efollett [otis
|
UW) Find the neqvestion Line of Yon X and hence predick ¥ 1h X=10-
W) FE a Regnesion Line @ x on Y ord hence pudich Xb Y=2-6
The actual meah are
o i do | dy? | doicky
1 6 2 ct cf 165 |e
5 \ 2 a 4 \ 2
3 ° ° 2 Oo a S
2 ° cl 2 ) 4 2
\ ) 2 a 4 1 2
1 2 ee ° 4 ° °
+ 1 4 +4 16 1 Py
3 {@ iS ° gq °
23 16 A ° EEA
The segresion ean & Yon xX &
y-¥ = byx (*- -X)
ee gone
_
gant (sad?
N
byxRegrersion egy y er X a +
(-2) = ~0-204 (x- 35
when x= 10
Y-2 = -0- 204 (10-3)
Y-2 = =0-304 @)
Y= 2-2-1298
Regression eqn & Xen ¥ &
Sel 2 N
“ 2
ray? day
1x0
= =o (ES) = 10
ag (Oo ee
Bo 3024
Regrermier ean of X on VU
t . C3) =-024(y-2)
when Y=9-S
X-3 = —0/29(2-S-2)
®-3 =-0.29 (0-8)
X= B+ 0-135