In: Statistics and Probability
Almost all U.S. light-rail systems use electric cars that run on
tracks built at street level. The Federal Transit Administration
claims light-rail is one of the safest modes of travel, with an
accident rate of .99 accidents per million passenger miles as
compared to 2.29 for buses. The following data show the miles of
track and the weekday ridership in thousands of passengers for six
light-rail systems.
City | Miles of Track | Ridership (1000s) |
Cleveland | 16 | 16 |
Denver | 18 | 36 |
Portland | 39 | 82 |
Sacramento | 22 | 32 |
San Diego | 48 | 76 |
San Jose | 32 | 31 |
St. Louis | 35 | 43 |
SSE | |
SST | |
SSR | |
MSE |
X | Y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
16 | 16 | 196.00 | 849.306 | 408.000 |
18 | 36 | 144.00 | 83.592 | 109.714 |
39 | 82 | 81.00 | 1358.449 | 331.714 |
22 | 32 | 64.00 | 172.735 | 105.143 |
48 | 76 | 324.00 | 952.163 | 555.429 |
32 | 31 | 4.00 | 200.020 | -28.286 |
35 | 43 | 25.00 | 4.592 | -10.714 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 210 | 316 | 838.000 | 3620.9 | 1471 |
mean | 30.00 | 45.14 | SSxx | SSyy | SSxy |
sample size , n = 7
here, x̅ = Σx / n= 30.00 ,
ȳ = Σy/n = 45.14
SSxx = Σ(x-x̅)² = 838.0000
SSxy= Σ(x-x̅)(y-ȳ) = 1471.0
a)
estimated slope , ß1 = SSxy/SSxx =
1471.0 / 838.000 =
1.8
intercept, ß0 = y̅-ß1* x̄ = -7.5
so, regression line is Ŷ = -7.5
+ 1.8 *x
b)
SSE= (SSxx * SSyy - SS²xy)/SSxx = 1038.708
SSt=SSyy = 3620.9
SSR=SST-SSE
MSE=SSE/(n-2)=207.7
SSE | 1038.7 |
SST | 3620.9 |
SSR | 2582.1 |
MSE=207.7 |
c)
R² = (Sxy)²/(Sx.Sy) = 0.7131
Yes, it provides a good fit
d)
X Value= 30
Confidence Level= 95%
Sample Size , n= 7
Degrees of Freedom,df=n-2 = 5
critical t Value=tα/2 = 2.571 [excel
function: =t.inv.2t(α/2,df) ]
X̅ = 30.00
Σ(x-x̅)² =Sxx 838.0
Standard Error of the Estimate,Se= 14.41
Predicted Y at X= 30 is
Ŷ = -7.518 + 1.755
* 30 = 45.143
standard error, S(ŷ)=Se*√(1/n+(X-X̅)²/Sxx) =
5.448
margin of error,E=t*Std error=t* S(ŷ) =
2.5706 * 5.4477 =
14.0037
Confidence Lower Limit=Ŷ +E = 45.143
- 14.0037 = 31.1
Confidence Upper Limit=Ŷ +E = 45.143
+ 14.0037 = 59.1
e)
standard error, S(ŷ)=Se*√(1+1/n+(X-X̅)²/Sxx) =
15.4084
margin of error,E=t*std error=t*S(ŷ)=
2.5706 * 15.41 =
39.6086
Prediction Interval Lower Limit=Ŷ -E =
45.143 - 39.6086 =
5.5
Prediction Interval Upper Limit=Ŷ +E =
45.143 + 39.6086 =
84.8