Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

MATHEMATICAL METHODS FOR COMPUTATION

(SESH2O51)

Tushil Mayani (22SE02IE019)


Abhijeet Savaliya (22SE02IE030)
Viraj Savaliya (22SE02IE031)
Pritkumar Vekariya (22SE02IE058)
Meet Vaghasiya (22SE02IE059)
REGRESSION ANALYSIS

• Covariance:
• Sample, Population
• Coefficient of correlation
C O VARIANCE
A measure of linear association between two variables. Positive values indicate
a positive relationship, negative values indicate a negative relationship.

Sample Covariance

Σ (x i − x̄ )(yi − ȳ)
C ov(x, y) = S x y =
n −1

Population Covariance

Σ(x i − µx )(yi − µy )
C ov(x, y) = σ x y =
N
Coefficient of Correlation

The numerical measure of correlation is called the coefficient of correlation and is


defined by the relation

r=
C ov(x , y)
= Σ (x − x̄ )(y − ȳ) Σ (x − x̄ )(y − ȳ)
=
√ var(x ) × var(y) nσ x σ y √ Σ (x − x̄ ) 2Σ(y − ȳ) 2
E x a m p l e : 10 student got the following marks of economy and statistics.
Roll No 1 2 3 4 5 6 7 8 9 10
Marks of Eco. 78 36 98 25 75 82 90 62 65 39 Calculate coefficient of Correlation
Marks of Stat. 84 51 91 60 68 62 86 58 53 37
Solution:
x y x − x̄ y − ȳ (x − x̄)2 (y − ȳ)2 (x − x̄)(y − ȳ)
78 84 13 19 169 361 247
36 51 -29 -14 841 196 406
98 91 33 26 1089 676 858
25 60 -40 -5 1600 25 200
Σ (x − x̄ )(y − ȳ) 75 68 10 3 100 9 30
r=
√ Σ (x − x̄ ) 2 Σ(y − ȳ) 2 82 62 17 -3 289 9 -51
90 86 25 21 625 441 525
62 58 -3 -7 9 49 21
65 53 0 -12 0 144 0
39 37 -26 -28 676 784 728
Σ = 650 Σ = 650 Σ = 0 Σ = 0 Σ = 5398 Σ = 2694 Σ = 2964
Σ (x − x̄ )(y − ȳ)
r=
√ Σ (x − x̄ ) 2 Σ (y − ȳ) 2

2964
=
√ 5398 × 2694

= 0.778
REGRESSION LINE
REGRESSION LINE

The equation of the line of regression of y on x is


σy
y − ȳ = r (x − x̄ )
σx
and The equation of the line of regression of x on y is
σx
x − x̄ = r (y − ȳ)
σy
σy
Thus the regression coefficient of y on x, by x = r and and the regression coefficient
σx
σx
of x on y, bxy = r Hence, r = ± √bx y .b y x
σy
Formula to find regression coefficient from data directly

nΣxy −ΣxΣy nΣxy −ΣxΣy


byx = or bxy =
nΣ y 2 − (Σy) 2 nΣ X 2 −(Σ x) 2
σy σx
by x = r or bx y = r
σx σy

C ov(x, y) C ov(x, y)
byx = or bxy =
(σ x ) 2 (σ y ) 2
EX AM P LE: OBTAIN REGRESSION LINE OF Y ON X
• FOR THE FOLLOWING DATA AND ESTIMATE Y CORRESPONDING
TO X = 0.6

x 1 2 3 4 5 6 7 8 9
y 9 8 10 12 11 13 14 16 15

• SOLUTION: BYX = 0.95, Y = 16.18


Σ Σ Σ Σ Σ
THANK YOU..

You might also like