Weight(X) | Price(Y) | XY | X² | Y² |
64 | 245 | 15680 | 4096 | 60025 |
64 | 275 | 17600 | 4096 | 75625 |
64 | 205 | 13120 | 4096 | 42025 |
64 | 201 | 12864 | 4096 | 40401 |
58 | 303 | 17574 | 3364 | 91809 |
47 | 709 | 33323 | 2209 | 502681 |
49 | 900 | 44100 | 2401 | 810000 |
59 | 331 | 19529 | 3481 | 109561 |
66 | 204 | 13464 | 4356 | 41616 |
58 | 308 | 17864 | 3364 | 94864 |
58 | 304 | 17632 | 3364 | 92416 |
52 | 489 | 25428 | 2704 | 239121 |
52 | 485 | 25220 | 2704 | 235225 |
63 | 361 | 22743 | 3969 | 130321 |
62 | 370 | 22940 | 3844 | 136900 |
54 | 565 | 30510 | 2916 | 319225 |
63 | 251 | 15813 | 3969 | 63001 |
63 | 279 | 17577 | 3969 | 77841 |
Ʃx = | Ʃy = | Ʃxy = | Ʃx² = | Ʃy² = |
1060 | 6785 | 382981 | 62998 | 3162657 |
Sample size, n = | 18 |
x̅ = Ʃx/n = 1060/18 = | 58.8888889 |
y̅ = Ʃy/n = 6785/18 = | 376.944444 |
SSxx = Ʃx² - (Ʃx)²/n = 62998 - (1060)²/18 = | 575.777778 |
SSyy = Ʃy² - (Ʃy)²/n = 3162657 - (6785)²/18 = | 605088.944 |
SSxy = Ʃxy - (Ʃx)(Ʃy)/n = 382981 - (1060)(6785)/18 = | -16580.111 |
Sum of Square error, SSE = SSyy -SSxy²/SSxx = 605088.94444 - (-16580.11111)²/575.77778 = 127647.66
Standard error, se = √(SSE/(n-2)) = √(127647.65535/(18-2)) = 89.31953
a) Scatter plot: Answer B.
b) There appears to be a negative linear relationship between variables.
--
c) Slope, b = SSxy/SSxx = -16580.11111/575.77778 = -28.79602
y-intercept, a = y̅ -b* x̅ = 376.94444 - (-28.79602)*58.88889 = 2072.7103
Regression equation :
Price = 2072.7 + (-28.8) Weight
--
For Intercept :
y-intercept, a = y̅ -b* x̅ = 376.94444 - (-28.79602)*58.88889 = 2072.71
Standard error for Intercept, se(b0) = se*√((1/n) + (x̅²/SSxx)) = 89.31953*√((1/18) + (58.88889²/575.77778)) = 220.2146
t = a /se(b0) = 9.412229
For slope :
Slope, b = SSxy/SSxx = -16580.11111/575.77778 = -28.796
Standard error for slope, se(b1) = se/√SSxx = 89.31953/√575.77778 = 3.7223652
t = b /se(b1) = -7.73595
Coefficients | Standard Error | t Stat | |
Constant | 2072.7 | 220.215 | 9.412 |
Weight | -28.8 | 3.722 | -7.736 |
-------
S = se = 89.3196
R Square = (SSxy)²/(SSxx*SSyy)
= (-16580.11111)²/(575.77778*605088.94444) = 0.7890 = 78.90%
---------
df(Regression) = 1
df(residual) = n-2 = 16
df(total) = n-1 = 17
SSR = SSxy²/SSxx = 477441.29
SSE = SSyy - SSxy²/SSxx =127647.66
SST = SSyy = Ʃy² - (Ʃy)²/n =605088.94
MSR = SSR/df(regression) = 477441.29
MSE = SSE/df(residual) = 7977.9785
F = MSR/MSE = 59.844896
p-value = F.DIST.RT(59.8449, 1, 16) = 0.0000
ANOVA | |||||
df | SS | MS | F | Significance F | |
Regression | 1 | 477441.3 | 477441.3 | 59.84 | 0.0000 |
Residual | 16 | 127647.7 | 7978.0 | ||
Total | 17 | 605088.9 |
p-value is less than 0.01
d) We can conclude that their is a significant relationship between variables.
e) Correlation coefficient, r = SSxy/√(SSxx*SSyy)
= -16580.11111/√(575.77778*605088.94444) = -0.8883
Automobile racing, high-performance driving schools, and driver education programs run by automobile clubs continue to grow...
To help consumers in purchasing a laptop computer, Consumer Reports calculates an overall test score for each computer tested based upon rating factors such as ergonomics, portability, performance, display, and battery life. Higher overall scores indicate better test results. The following data show the average retail price and the overall score for ten 13-inch models (Consumer Reports website, October 25, 2012). Click on the datafile logo to reference the data DATA file Brand & Model Samsung Ultrabook NP900X3C-A0IUS Apple MacBook...
To help consumers in purchasing a laptop computer, Consumer Reports calculates an overall test score for each computer tested based upon rating factors such as ergonomics, portability, performance, display, and battery life. Higher overall scores indicate better test results. The following data show the average retail price and the overall score for ten 13-inch models (Consumer Reports website, October 25, 2012). To help consumers in purchasing a laptop computer, Consumer Reports calculates an overall test score for each computer tested...