In: Statistics and Probability
Use the data in the table below to complete parts (a) through (c).
X: 5 5 6 9 13 16 18 48
Y: 30 27 18 23 25 20 21 10
A.) Construct a scatterplot of the data.
B.) Indentify any possible outliners.
A.) The point (18,21) may be an outliner.
B.) The point (48,10) may be an outliner
C.) The point (5,27) may be an outliner.
D.) The point (5,30) may be an outliner
E.) There are no outliners.
C.) Determine if the point is influential. The change in slope or intercept is significant if it larger than 10%
The point is/is not an influential point because the slopes with the point included and without the point included are/are not signficatly different, and the intercepts are/ are not significantly different.
a)
B)
The point (48,10) may be an outliner
C)
with outlier slope is
X | Y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
5 | 30 | 100.000 | 68.063 | -82.500 |
5 | 27 | 100.000 | 27.563 | -52.500 |
6 | 18 | 81.000 | 14.063 | 33.750 |
9 | 23 | 36.000 | 1.563 | -7.500 |
13 | 25 | 4.000 | 10.563 | -6.500 |
16 | 20 | 1.000 | 3.063 | -1.750 |
18 | 21 | 9.000 | 0.563 | -2.250 |
48 | 10 | 1089.000 | 138.063 | -387.750 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 120 | 174 | 1420.000 | 263.5 | -507 |
mean | 15.00 | 21.75 | SSxx | SSyy | SSxy |
estimated slope , ß1 = SSxy/SSxx = -507.0 / 1420.000 = -0.3570
intercept, ß0 = y̅-ß1* x̄ =
27.1056
--------------------------
without outlier slope is,
X | Y | (x-x̅)² | (y-ȳ)² | (x-x̅)(y-ȳ) |
5 | 30 | 27.939 | 43.184 | -34.735 |
5 | 27 | 27.939 | 12.755 | -18.878 |
6 | 18 | 18.367 | 29.469 | 23.265 |
9 | 23 | 1.653 | 0.184 | 0.551 |
13 | 25 | 7.367 | 2.469 | 4.265 |
16 | 20 | 32.653 | 11.755 | -19.592 |
18 | 21 | 59.510 | 5.898 | -18.735 |
ΣX | ΣY | Σ(x-x̅)² | Σ(y-ȳ)² | Σ(x-x̅)(y-ȳ) | |
total sum | 72 | 164 | 175.429 | 105.7 | -63.8571429 |
mean | 10.29 | 23.43 | SSxx | SSyy | SSxy |
estimated slope , ß1 = SSxy/SSxx = -63.9 / 175.429 = -0.3640
intercept, ß0 = y̅-ß1* x̄ =
27.1726
Change in slope = (-0.364+0.357)/-0.357 = 1.96%
change in intercept = (27.1726-27.1056)/27.1056 = 0.2%
The point is not an influential point because the slopes with the point included and without the point included are not signficatly different, and the intercepts are not significantly different.