In: Statistics and Probability
Listed below are amounts of court income and salaries paid to the town justices for a certain town. All amounts are in thousands of dollars. Find the (a) explained variation, (b) unexplained variation, and (c) indicated prediction interval. There is sufficient evidence to support a claim of a linear correlation, so it is reasonable to use the regression equation when making predictions. For the prediction interval, use a 99% confidence level with a court income of $800,000.
Court Income $60 $408 $1583 $1124 $268 $252 $108 $169 $29
Justice Salary $26 $40 $96 $60 $46 $57 $30 $27 $15
A) Find explained variation. round to 3 decimal places
B) Find unexplained variation. round to 3 decimal places
C) Find indicated prediction interval. round to 3 decimal places
X | Y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
60 | 26 | 147882.98 | 328.01 | 6964.73 |
408 | 40 | 1336.31 | 16.90 | 150.28 |
1583 | 96 | 1296055.75 | 2692.46 | 59072.62 |
1124 | 60 | 461644.75 | 252.46 | 10795.62 |
268 | 46 | 31171.86 | 3.57 | -333.49 |
252 | 57 | 37077.64 | 166.12 | -2481.83 |
108 | 30 | 113269.64 | 199.12 | 4749.17 |
169 | 27 | 75930.86 | 292.79 | 4715.06 |
29 | 15 | 172686.42 | 847.46 | 12097.28 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 4001 | 397 | 2337056.222 | 4798.9 | 95729.44 |
mean | 444.56 | 44.11 | SSxx | SSyy | SSxy |
a)
explained variation=R² = (Sxy)²/(Sx.Sy) = 0.817
b)
Unexplained variation= 1- 0.817 =0.183
c)
sample size , n = 9
here, x̅ = Σx / n= 444.56 ,
ȳ = Σy/n = 44.11
SSxx = Σ(x-x̅)² =
2337056.2222
SSxy= Σ(x-x̅)(y-ȳ) = 95729.4
estimated slope , ß1 = SSxy/SSxx = 95729.4
/ 2337056.222 = 0.0410
intercept, ß0 = y̅-ß1* x̄ =
25.9014
so, regression line is Ŷ =
25.901 + 0.040962 *x
SSE= (SSxx * SSyy - SS²xy)/SSxx =
877.6625
std error ,Se = √(SSE/(n-2)) =
11.19734
--------------
X Value= 800 (in thousands of dollars)
Confidence Level= 99%
Sample Size , n= 9
Degrees of Freedom,df=n-2 = 7
critical t Value=tα/2 = 3.499 [excel
function: =t.inv.2t(α/2,df) ]
X̅ = 444.56
Σ(x-x̅)² =Sxx 2337056.2
Standard Error of the Estimate,Se= 11.197
Predicted Y at X= 800 is
Ŷ = 25.901 + 0.041
* 800 = 58.67
For Individual Response Y
standard error, S(ŷ)=Se*√(1+1/n+(X-X̅)²/Sxx) =
12.0868
margin of error,E=t*std error=t*S(ŷ)=
3.4995 * 12.09 =
42.2974
Prediction Interval Lower Limit=Ŷ -E =
58.671 - 42.30 =
16.373 (in thousands of dollars)
Prediction Interval Upper Limit=Ŷ +E =
58.671 + 42.30 =
100.968 (in thousands of dollar)