In: Statistics and Probability
A random sample of 20 workers in a factory were asked to report the age of their car and how many miles the vehicle had on it. A computer printout resulted in the following information.
Variable |
Coef |
SE Coef |
t-ratio |
Prob |
Constant |
7288.54 |
6591 |
1.11 |
<0.2826 |
Age |
11630.6 |
1249 |
9.31 |
<0.0001 |
R sq = 82% |
R sq adj = 81.1% |
s = 19280 |
The given result a simple linear model to predict age of the car.
The simple linear regression is used to predict a continuous outcome variable (y) based on one single predictor variable (x).
The mathematical formula is
y = b0 + b1*x +
e
Which means:
beta1 (b1) times x, plus a
constant beta0 (b0), plus an error term e
· b0 is the intercept
· b1 is the slope of the regression line
· e is the error term
From the give computer printed results,
x = Age of the car
y = miles the vehicle had on car
Intercept = Coef
The intercept (b0) is 7288.54 and coefficient for Age variable (b1), also known as the slope, is 11630
The following is the estimated regression equation:
Miles = 7288.54 + 11630 * Age of the car
If the age of the car is given then using this formula we can calculate its miles.
Question1: Find the LSRL
Answer: A regression line (LSRL - Least Squares Regression Line) is a straight line that describes how a response variable y changes as an explanatory variable x changes.
To find LSRL:
Question2: A new worker starts next week and we know that his car is 7 years old, how many miles would you expect to be on his car?
Answer: Miles = 7288.54 + (11630 * Age of the car)
Miles = 7288.54 + 11630*7 = 88698.54
miles would you expect to be on his car is 88698.54
Question3: Interpret the slope of the LSRL in the context of this problem.
Answer:
For a car with age7 we can expect an increase of 81410 (i.e., 11630 *7) miles
That is, Miles = 7288.54 + 11630*7 = 88698.54
Question4: Find a 95% confidence interval for the slope of the LSRL
Answer:
Slope b1 = 11630
Confidence level = 95%
Compute alpha (α):
α = 1 - (confidence level / 100) = 1 - (95/100) = 0.05
Find the critical probability (p):
P = 1 - α/2 = 1 - 0.05/2 = 0.975
Find the degrees of freedom (df):
Here n = 20 (sample of 20 workers)
df = n - 2 = 20 - 2 = 18.
Find critical value:
From the T distribution critical value table we can identify critical value for degrees of freedom = 18 and α = 0.05
critical value = 1.734
Compute margin of error (ME):
ME = critical value * standard error
SE Coef in the model represents the standard error.
ME = 1.734 * 1249 = 2165.766
The range of the confidence interval is defined by the sample statistic + or - margin of error
Therefore, the 95% confidence interval for this sample is 11630 + or - 2165.766
85% confidence interval
Confidence level = 85%
α = 1 - (confidence level / 100) = 1 - (85/100) = 0.15
Degrees of freedom (df) = n-2 = 20 – 2 = 18
Critical value = 1.504
ME = critical value * standard error = 1.504 * 1249 = 1878.496
Therefore, the 85% confidence interval for this sample is 11630 + or - 1878.496