Scribd
Upload
2 views

TEST of correlation

The document presents a calculation of the sample correlation coefficient, which is found to be -0.975 based on given sample data. It also outlines a hypothesis test for the correlation with null hypothesis H0: ρ=0 and alternative hypothesis H1: ρ>0, resulting in a test statistic and p-value. Finally, it concludes that the sample correlation coefficient must be 0.631 under the specified test conditions.

Uploaded by

sarthakgarg0401
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views

TEST of correlation

The document presents a calculation of the sample correlation coefficient, which is found to be -0.975 based on given sample data. It also outlines a hypothesis test for the correlation with null hypothesis H0: ρ=0 and alternative hypothesis H1: ρ>0, resulting in a test statistic and p-value. Finally, it concludes that the sample correlation coefficient must be 0.631 under the specified test conditions.

Uploaded by

sarthakgarg0401
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

TEST of correlation

S xy
Ans 1. sample correlation coefficient= 1

( S xx S yy ) 2
for 200 samples∈form of ( x , y )
(−2 , 2 ) occurs50׿
( 0 , 0 ) occurs 90׿
( 2 ,−1 ) occurs 60׿

then ,∑ X i Y i =50 ( (−2 )( 2 )) + 90 ( ( 0 ) ( 0 ) ) +60 ( ( 2 )(−1 ) )

¿−320
∑ X i=50 (−2 ) +90 ( 0 ) +60 ( 2 )

¿ 20
∑ Y i=50 ( 2 ) + 90 ( 0 ) +60 (−1 )

¿ 40

∑ X 2i =50 (−2 ) + 90 ( 02 ) +60 ( 22 )


2

¿ 440

∑ Y 2i =50 ( 22 ) + 90 ( 02 ) +60 (−12 )


¿ 260

∑ X ∑Y ( 20 ( 40 ) )
hence , S xy =∑ XY − =−320− =−324
n 200

2 ( ∑ X )2 ( 20 2 )
S xx =∑ X − =440− =438
n 200

( ∑ Y )2 402
S yy =∑ Y 2− =260− =252
n 200
−324
therefore , r= 1
=−0.975
( 438∗252 ) 2

so , required correlation coefficient=−0.975


Ans 2. H 0 : ρ=0

H 1 : ρ>0

under H 0 test statistics will be


r ( √ n−2 )
t n−2
√1−r 2
if n=10

r ( √ n−2 ) r ( √ 10−2 )
then ,test statistics value= =
√ 1−r 2
√ 1−r 2
¿ , for one tailed test p−value is

(
p t n−2 >
r ( √10−2 )
√ 1−r 2 ) =0.025

( √
( (√
p t 8>
r 10−2 )
1−r
2 ))
=0.025

¿ actuarial table page 163

r ( √10−2 )
=2.306
√ 1−r 2

√
1
then ,r = =0.631
8
1+ 2
( 2.306 )
hence , sample correlation coefficient must be 0.631

You might also like