In: Statistics and Probability
An experiment was performed on a certain metal to determine if the strength is a function of heating time (hours). Results based on 25 metal sheets are given below. Use the simple linear regression model.
∑X = 50
∑X2 = 200
∑Y = 75
∑Y2 = 1600
∑XY = 400
Find the estimated y intercept and slope. Write the equation of the least squares regression line and explain the coefficients. Estimate Y when X is equal to 4 hours. Also determine the standard error, the Mean Square Error, the coefficient of determination and the coefficient of correlation. Check the relation between correlation coefficient and Coefficient of Determination. Test the significance of the slope.
Please show all work, please type out so it is legible, thank you
Answer:
Given,
Ʃ x = 50
Ʃ x² = 200
Ʃ y = 75
Ʃ y² = 1600
Ʃ xy = 400
Sample size, n = 25
Now let us consider,
x̅ = Ʃx/n
substitute values
= 50/25
x̅ = 2
y̅ = Ʃy/n
substitute values
= 75/25
y̅ = 3
Now,
SSxx = Ʃx² - (Ʃx)²/n
substitute values in above formula
= 200 - (50)²/25
= 200 - 100
SSxx = 100
SSyy = Ʃy² - (Ʃy)²/n
substitute values in above formula
= 1600 - (75)²/25
= 1600 - 225
SSyy = 1375
SSxy = Ʃxy - (Ʃx)(Ʃy)/n
substitute values in above formula to get required value
= 400 - (50)(75)/25
= 400 - 150
SSxy = 250
Slope, b = SSxy/SSxx
substitute values to get slope
= 250/100
b = 2.5
consider,
y-intercept,
a = y̅ - b* x̅
= 3 - (2.5)*2
= 3 - 5
a = - 2
Now we have to consider the regression equation :
ŷ = - 2 + (2.5)*x
substitute x = 4 in above formula
So the predicted value of y is given below
ŷ = - 2 + (2.5) * 4
= - 2 + 10
ŷ = 8
SSE = SSyy - SSxy²/SSxx
substitute the known values
= 1375 - (250)²/100
= 1375 - 625
SSE = 750
Standard error = √(SSE/(n-2))
= √(750/(25 - 2))
= √32.61
Standard error = 5.71040
Mean square error
MSE = SSE/(n-2)
= 750/(25-2)
= 750/23
MSE = 32.6087
Correlation coefficient, r = SSxy/√(SSxx*SSyy)
= 250/√(100*1375)
= 250/370.81
r = 0.6742
Coefficient of determination,
r² = (SSxy)²/(SSxx*SSyy)
= (250)²/(100*1375)
= 62500/137500
r² = 0.4545
Correlation coefficient is given as the square root of coefficient of determination
Slope Hypothesis test: can be given as follows
Null hypothesis
Ho: β₁ = 0
Alternative hypothesis
Ha: β₁ ≠ 0
Consider test statistic:
t = b / (SE/√SSxx)
now after substituting the values and then we get
t = 4.3780
degree of freedom = n - 2
= 25 - 2
df = 23
Critical value,
Corresponding t value = t(0.05, 23)
t = 2.067
P value = corresponding p value for t(4.378), 23 is 0.0002
So
P value = 0.0002
Here we observe that p-value < alpha , so we reject the null hypothesis Ho.