In: Statistics and Probability
Sales of Body Price of Lens Gender Sales of Lens
155 $700 1 122
101 650 1 120
157 725 0 135
180 575 1 95
150 600 0 100
201 750 0 174
99 560 1 118
137 500 0 130
155 675 1 128
165 550 1 166
152 725 0 131
127 750 1 102
217 565 0 165
186 670 0 154
176 600 1 97
123 585 0 129
109 645 0 98
90 575 1 105
176 660 0 120
129 650 1 105
f) Find the sample coefficient of determination for the data. Explain its meaning in the context of the problem.
g) Find the adjusted sample coefficient of determination for the data. Explain its meaning in the context of the problem. Why does it differ, if it does, from the coefficient of determination that you found in the previous part of the problem?
We first run a regression analysis on the given data in minitab-
Steps for regression-
1. Enter the given values in different columns.
2. Go to “Stat” then “Regression” then “Regression” then “Fit regression model”.
3. Enter Sales of body in “response” and price of leens,gender,sales of lens in “continuous predictor”.
4. Click OK.
After running the above steps we get the following output-
Model Summary
S R-sq R-sq(adj) R-sq(pred)
31.0886 33.07% 20.52% 2.52%
f) The sample coefficient of determination for the data is 0.3307.
We know that coefficient of determination is a measure of goodness of fit for regression model. This statistic also interprets the percentage of variation in the dependent variable which is explained by the independent variable. According to our value we can say that it is not a good fit.
g) The adjusted sample coefficient of determination for the data is 0.2052.
The adjusted coefficient of determination value increases or decreases compared to the coefficient of determination if a new term improves or doesnot improve the fit of the model respectively.
Here, our value is less trhan the sample coefficient of determination so we can say that there is no improvement in the model.
The adjusted coefficient of determination adjusts for the number of terms in the regression model. Here we see that both the values are different from each other because the adjusted sample coefficient of determination adjusts for the number of terms in the model which the normal sample coefficient of determination doesn't.