In: Statistics and Probability
you must calculate the deviations from the average, the manual correlation calculation and cross check it with Excel's correl function. Furthermore, you will generate the scatterplot graph, along with the trendline. Indicate whether there is a strong/weak/no correlation between the two variables.
Question:
The following tables shows the selling prices, in thousands of dollars, and the square footages of seven randomly selected homes recently sold by Century 21 Realtors:
Price: 258 191 253 168 249 245 282
Sqr Ft: 2730 1860 2140 2180 2310 2450 2920
Using XLS, calculate the correlation coefficient for this sample. Using an significance of .1, test the significance of the population correlation coefficient between a house’s selling price and its square footage. What conclusions can you draw?
( X) | ( Y) | X^2 | Y^2 | X*Y |
258 | 2730 | 66564 | 7452900 | 704340 |
191 | 1860 | 36481 | 3459600 | 355260 |
253 | 2140 | 64009 | 4579600 | 541420 |
168 | 2180 | 28224 | 4752400 | 366240 |
249 | 2310 | 62001 | 5336100 | 575190 |
245 | 2450 | 60025 | 6002500 | 600250 |
282 | 2920 | 79524 | 8526400 | 823440 |
calculation procedure for correlation
sum of (x) = ∑x = 1646
sum of (y) = ∑y = 16590
sum of (x^2)= ∑x^2 = 396828
sum of (y^2)= ∑y^2 = 40109500
sum of (x*y)= ∑x*y = 3966140
to caluclate value of r( x,y) = covariance ( x,y ) / sd (x) * sd
(y)
covariance ( x,y ) = [ ∑x*y - N *(∑x/N) * (∑y/N) ]/n-1
= 3966140 - [ 7 * (1646/7) * (16590/7) ]/7- 1
= 9302.857
and now to calculate r( x,y) = 9302.857/
(SQRT(1/7*3966140-(1/7*1646)^2) ) * (
SQRT(1/7*3966140-(1/7*16590)^2)
=9302.857 / (37.384*336.197)
=0.74
value of correlation is =0.74
coefficient of determination = r^2 = 0.548
properties of correlation
1. If r = 1 Corrlation is called Perfect Positive Correlation
2. If r = -1 Correlation is called Perfect Negative
Correlation
3. If r = 0 Correlation is called Zero Correlation
& with above we conclude that correlation ( r ) is = 0.7402>
0 ,perfect positive correlation
Given that,
value of r =0.74
number (n)=7
null, Ho: ρ =0
alternate, H1: ρ!=0
level of significance, α = 0.1
from standard normal table, two tailed t α/2 =2.015
since our test is two-tailed
reject Ho, if to < -2.015 OR if to > 2.015
we use test statistic (t) = r / sqrt(1-r^2/(n-2))
to=0.74/(sqrt( ( 1-0.74^2 )/(7-2) )
to =2.46
|to | =2.46
critical value
the value of |t α| at los 0.1% is 2.015
we got |to| =2.46 & | t α | =2.015
make decision
hence value of | to | > | t α| and here we reject Ho
ANSWERS
---------------
null, Ho: ρ =0
alternate, H1: ρ!=0
test statistic: 2.46
critical value: -2.015 , 2.015
decision: reject Ho
we have enough evidence to support the claim that the significance
of the population correlation coefficient between a house’s selling
price and its square footage.