In: Statistics and Probability
Given the following pairs of data where x is the number of years of experience in a certain company and y is the annual salary:
Determine the equation of the linear regression line. Round the slope and y-intercept to the nearest integers.
Using your answer in (a), estimate the starting salary for a new hire.
Using your answer in (a), estimate the number of years (to the nearest year) an employee would have to work in the company in order to earn an annual salary of $175,000
X Y
8 70,000
10 100,000
12 115,000
13 120,000
15 120,000
16 125,000
17 130,000
18 140,000
19 145,000
20 150,000
X | Y | XY | X² | Y² |
8 | 70000 | 560000 | 64 | 4900000000 |
10 | 100000 | 1000000 | 100 | 10000000000 |
12 | 115000 | 1380000 | 144 | 13225000000 |
13 | 120000 | 1560000 | 169 | 14400000000 |
15 | 120000 | 1800000 | 225 | 14400000000 |
16 | 125000 | 2000000 | 256 | 15625000000 |
17 | 130000 | 2210000 | 289 | 16900000000 |
18 | 140000 | 2520000 | 324 | 19600000000 |
19 | 145000 | 2755000 | 361 | 21025000000 |
20 | 150000 | 3000000 | 400 | 22500000000 |
Ʃx = | 148 |
Ʃy = | 1215000 |
Ʃxy = | 18785000 |
Ʃx² = | 2332 |
Ʃy² = | 152575000000 |
Sample size, n = | 10 |
x̅ = Ʃx/n = 148/10 = | 14.8 |
y̅ = Ʃy/n = 1215000/10 = | 121500 |
SSxx = Ʃx² - (Ʃx)²/n = 2332 - (148)²/10 = | 141.6 |
SSyy = Ʃy² - (Ʃy)²/n = 152575000000 - (1215000)²/10 = | 4952500000 |
SSxy = Ʃxy - (Ʃx)(Ʃy)/n = 18785000 - (148)(1215000)/10 = | 803000 |
a)
Slope, b = SSxy/SSxx = 803000/141.6 = 5670.903955
y-intercept, a = y̅ -b* x̅ = 121500 - (5670.90395)*14.8 = 37570.62147
Regression equation :
ŷ = 37571 + (5671) x
b)
Predicted value of y for a new hire whose experience is 0
ŷ = 37571 + (5671) * 0 = 37571
c)
if salary = 175000 then number of years of experience =
175000 = 37571 + (5671) x
x = 24.23 = 24 years