In: Statistics and Probability
Multiple regression analysis was used to study the relationship between a dependent variable, y, and four independent variables; x1, x2, x3, and x4. The following is a partial result of the regression analysis involving 31 observations.
Coefficients |
Standard Error |
|||
Intercept |
18.00 |
6.00 |
||
x1 |
12.00 |
8.00 |
||
x2 |
24.00 |
48.00 |
||
x3 |
-36.00 |
36.00 |
||
x4 |
16.00 |
2.00 |
||
|
||||
df |
SS |
MS |
F |
|
Regression |
125 |
|||
Error |
||||
Total |
760 |
a) Compute the multiple coefficient of determination.
b) Perform a t test and determine whether or not β1 is significantly different from zero (α = .05).
c) Perform a t test and determine whether or not β4 is significantly different from zero (α = .05).
d) At α = .05, perform an F test and determine whether or not the regression model is significant.
Solution: To find the answers to part a through d, we need to complete the given tables. The tables are filled as:
ANOVA:
The p values are found using the t distribution for
a) Compute the multiple coefficient of determination.
Answer: The multiple coefficient of determination is given below:
rounded to four decimal places
b) Perform a t test and determine whether or not β1 is significantly different from zero (α = .05).
Answer: The null and the alternative hypotheses are:
The test statistic from the above table is:
And the is:
Since the p-value is greater than the significance level 0.05, therefore we fail to reject the null hypothesis and conclude that β1 is not significantly different from zero
c) Perform a t test and determine whether or not β4 is significantly different from zero (α = .05).
Answer: The null and the alternative hypotheses are:
The test statistic from the above table is:
And the is:
Since the p-value is less than the significance level 0.05, therefore we reject the null hypothesis and conclude that β4 is significantly different from zero.
d) At α = .05, perform an F test and determine whether or not the regression model is significant.
Answer: The null and the alternative hypotheses are:
At least one coefficient is different from 0.
The test statistic from the above ANOVA table is:
And the using the F distribution for is:
Since the p-value is less than the significance level 0.05, therefore we reject the null hypothesis and conclude at least one coefficient is significantly different from zero.