In: Statistics and Probability
The following data show the brand, price ($), and the overall
score for 6 stereo headphones that were tested by Consumer
Reports. The overall score is based on sound quality and
effectiveness of ambient noise reduction. Scores range from 0
(lowest) to 100 (highest). The estimated regression equation for
these data is = 23.395 + 0.316 x, where
x = price ($) and y = overall score.
Brand | Price ($) | Score |
Bose | 180 | 75 |
Scullcandy | 150 | 72 |
Koss | 95 | 61 |
Phillips/O'Neill | 70 | 56 |
Denon | 70 | 40 |
JVC | 35 | 26 |
Round your answers to three decimal places.
a. Compute SST, SSR, and SSE.
SST | |
SSR | |
SSE |
b. Compute the coefficient of determination r 2. Comment on the goodness of fit.
r 2 =
SelectThe least squares line provided a very good fitThe least squares line provided a bad fitItem 5
c. What is the value of the sample correlation coefficient?
r =
Line of Regression Y on X i.e Y = bo + b1 X | ||||
X | Y | (Xi - Mean)^2 | (Yi - Mean)^2 | (Xi-Mean)*(Yi-Mean) |
180 | 75 | 6400 | 400 | 1600 |
150 | 72 | 2500 | 289 | 850 |
95 | 61 | 25 | 36 | -30 |
70 | 56 | 900 | 1 | -30 |
70 | 40 | 900 | 225 | 450 |
35 | 26 | 4225 | 841 | 1885 |
calculation procedure for regression
mean of X = ∑ X / n = 100
mean of Y = ∑ Y / n = 55
∑ (Xi - Mean)^2 = 14950
∑ (Yi - Mean)^2 = 1792
∑ (Xi-Mean)*(Yi-Mean) = 4725
b1 = ∑ (Xi-Mean)*(Yi-Mean) / ∑ (Xi - Mean)^2
= 4725 / 14950
= 0.316
bo = ∑ Y / n - b1 * ∑ X / n
bo = 55 - 0.316*100 = 23.395
value of regression equation is, Y = bo + b1 X
Y'=23.395+0.316* X
a.
( X) | ( Y) | X^2 | Y^2 | X*Y |
180 | 75 | 32400 | 5625 | 13500 |
150 | 72 | 22500 | 5184 | 10800 |
95 | 61 | 9025 | 3721 | 5795 |
70 | 56 | 4900 | 3136 | 3920 |
70 | 40 | 4900 | 1600 | 2800 |
35 | 26 | 1225 | 676 | 910 |
calculation procedure for correlation
sum of (x) = ∑x = 600
sum of (y) = ∑y = 330
sum of (x^2)= ∑x^2 = 74950
sum of (y^2)= ∑y^2 = 19942
sum of (x*y)= ∑x*y = 37725
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
= 37725 - [ 6 * (600/6) * (330/6) ]/6- 1
= 787.5
and now to calculate r( x,y) = 787.5/ (SQRT(1/6*37725-(1/6*600)^2)
) * ( SQRT(1/6*37725-(1/6*330)^2)
=787.5 / (49.917*17.282)
=0.913
value of correlation is =0.913
b.
coeffcient of determination = r^2 = 0.833
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 = 0.9129>
0 ,perfect positive correlation
c.
value of correlation is =0.913