EXAMPLE 11.24 Square Feet, Bedrooms, and Bathrooms

homes04

CASE 11.3 Figure 11.21 gives the output for predicting price using SqFt, Bed3, B2, and Bh. The overall model is statistically significant (, , ), and it explains 58.9% of the variation in price.

image
Figure 11.21: FIGURE 11.21 Output for predicting price using square feet, bedroom, and bathroom information, Example 11.24.

The individual for Bed3 is not statistically significant (, , ). That is, in a model that contains square feet and information about the bathrooms, there is no additional information in the number of bedrooms that is useful for predicting price. This happens because the explanatory variables are related to each other: houses with more bedrooms tend to also have more square feet and more baths.

Therefore, we redo the regression without Bed3. The output appears in Figure 11.22. The value of has decreased slightly to 57.7%, but now all the coefficients for the explanatory variables are statistically significant. The fitted regression equation is

577

image
Figure 11.22: FIGURE 11.22 Output for predicting price using square feet and bathroom information, Example 11.24.