In: Statistics and Probability
Given the following sample observations, draw a scatter diagram on a separate piece of paper.
X: | −7 | −16 | 13 | 3 | 16 |
Y: | 54 | 241 | 157 | 1 | 342 |
Click here for the Excel Data File
Compute the correlation coefficient. (Round your answer to 3 decimal places.)
Does the relationship between the variables appear to be linear?
Yes
No
Try squaring the x variable and then determine the correlation coefficient. (Round your answer to 3 decimal places.)
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 9 | 795 | 722.8 | 76206.0 | 1851.00 |
mean | 1.80 | 159.00 | SSxx | SSyy | SSxy |
sample size , n = 5
here, x̅ = Σx / n= 1.80 ,
ȳ = Σy/n = 159.00
SSxx = Σ(x-x̅)² = 722.8000
SSxy= Σ(x-x̅)(y-ȳ) = 1851.0
estimated slope , ß1 = SSxy/SSxx = 1851.0
/ 722.800 = 2.5609
intercept, ß0 = y̅-ß1* x̄ =
154.3904
so, regression line is Ŷ =
154.3904 + 2.5609 *x
SSE= (SSxx * SSyy - SS²xy)/SSxx =
71465.822
std error ,Se = √(SSE/(n-2)) =
154.344
correlation coefficient , r = Sxy/√(Sx.Sy)
= 0.249
2)
Yes but weak positive
3)
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 739 | 795 | 52890.8 | 76206.0 | 60935.00 |
mean | 147.80 | 159.00 | SSxx | SSyy | SSxy |
sample size , n = 5
here, x̅ = Σx / n= 147.80 ,
ȳ = Σy/n = 159.00
SSxx = Σ(x-x̅)² = 52890.8000
SSxy= Σ(x-x̅)(y-ȳ) = 60935.0
estimated slope , ß1 = SSxy/SSxx = 60935.0
/ 52890.800 = 1.1521
intercept, ß0 = y̅-ß1* x̄ =
-11.2790
so, regression line is Ŷ =
-11.2790 + 1.1521 *x
SSE= (SSxx * SSyy - SS²xy)/SSxx =
6003.352
std error ,Se = √(SSE/(n-2)) =
44.734
correlation coefficient , r = Sxy/√(Sx.Sy)
= 0.96
THANKS
revert back for doubt
please upvote