In: Math
X | Y | XY | X² | Y² |
2292 | 393.8 | 902589.6 | 5253264 | 155078.44 |
3216 | 381.5 | 1226904 | 10342656 | 145542.25 |
1074 | 181.5 | 194931 | 1153476 | 32942.25 |
1948 | 333.8 | 650242.4 | 3794704 | 111422.44 |
3196 | 634 | 2026264 | 10214416 | 401956 |
2670 | 354 | 945180 | 7128900 | 125316 |
4126 | 629.7 | 2598142.2 | 17023876 | 396522.09 |
2126 | 363.8 | 773438.8 | 4519876 | 132350.44 |
2637 | 429.4 | 1132327.8 | 6953769 | 184384.36 |
1707 | 298.1 | 508856.7 | 2913849 | 88863.61 |
1855 | 281.9 | 522924.5 | 3441025 | 79467.61 |
3930 | 708.6 | 2784798 | 15444900 | 502113.96 |
Ʃx = | Ʃy = | Ʃxy = | Ʃx² = | Ʃy² = |
30777 | 4990.1 | 14266599 | 88184711 | 2355959.45 |
Sample size, n = | 12 |
x̅ = Ʃx/n = 30777/12 = | 2564.75 |
y̅ = Ʃy/n = 4990.1/12 = | 415.841667 |
SSxx = Ʃx² - (Ʃx)²/n = 88184711 - (30777)²/12 = | 9249400.25 |
SSyy = Ʃy² - (Ʃy)²/n = 2355959.45 - (4990.1)²/12 = | 280867.949 |
SSxy = Ʃxy - (Ʃx)(Ʃy)/n = 14266599 - (30777)(4990.1)/12 = | 1468240.03 |
a) Correlation coefficient, r = SSxy/√(SSxx*SSyy)
= 1468240.025/√(9249400.25*280867.94917) = 0.9109
b) Slope, b = SSxy/SSxx = 1468240.025/9249400.25 = 0.158738944
y-intercept, a = y̅ -b* x̅ = 415.84167 - (0.15874)*2564.75 = 8.715960963
Regression equation :
ŷ = 8.716 + (0.1587) x
c) For every additional square foot, the selling price increased by 0.1587 thousand dollars, on average
d) No, it is not reasonable to interpret the y-intercept.
A house of 0 square feet is not possible and outside the scope of the model.
e) Predicted value of y at x = 1431
ŷ = 8.716 + (0.1587) * 1431 = $ 235.87
The average price of a home that is 1431 feet is
$235.87 thousand.
This home’s price is below average for a home of this size.