{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Lecture 6 lecture notes

# Lecture 6 lecture notes - 0.00715-Signif codes 0 0.001 0.01...

This preview shows page 1. Sign up to view the full content.

Lecture 6. Regression with Dummy Variables Example . Again we consider the Albuquerque Home Prices. Now we shall add one more regressor, namely custom build or not, to our MLR model of prices vs. square footage and age. Information about a type of house is coded as 1 for custom build and 0 otherwise. > l<-lm(data\$PRICE~data\$SQFT+data\$AGE+data\$CUST) > summary(l) Call: lm(formula = 100 * data\$PRICE ~ data\$SQFT + data\$AGE + data\$CUST) Residuals: Min 1Q Median 3Q Max -56871.580 -9368.399 2.080 8506.354 63302.217 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 7248.077 8452.629 0.857 0.39437 data\$SQFT 63.914 4.712 13.565 < 2e-16 *** data\$AGE -428.913 168.111 -2.551 0.01313 * data\$CUST 14931.462 5372.717 2.779
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 0.00715 **---Signif. codes: 0 *** 0.001 ** 0.01 * 0.05 . 0.1 1 Residual standard error: 17420 on 64 degrees of freedom (49 observations deleted due to missingness) Multiple R-squared: 0.8216, Adjusted R-squared: 0.8132 F-statistic: 98.22 on 3 and 64 DF, p-value: < 2.2e-16 The new regressor CUST is found to be statistically signiﬁcant with a p-value of 0.007. The obtained adjusted R 2 of 81% is also slightly higher than for the ”old” MLR model with 79% for R 2 . The new MLR takes the form PRICE = 7248 . 077 + 63 . 914SQFT-428 . 913AGE + 14931CUST + ². (6.1) Hence, we can expect to pay about \$14,931 more for a custom build house. 1...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online