Question

Data from a random sample of 220 house sales.

Let Price denotes selling price in $1000, BDR denotes the number of bedrooms, Bath

denotes the number of bathrooms, Hsize denotes the size of house in square feet, Lsize

denotes the lot size in square feet, Age denotes the age of the house in years and Poor

denotes a binary variable that is equal to 1 if the condition of the house is reported

as “poor”. Standard errors in parenthesis.


119.2 + 0.485BDR + 23.4Bath + 0.156Hsize + 0.002Lsize + 0.090Age − 48.8Poor,

(23.9) (2.61) (8.94) (0.011) (0.00048) (0.311) (10.5)


1.The p-value is equal to 0.853 for testing the null hypothesis that the coefficient on

BDR is zero. Is that coefficient significantly different from zero at the 5% level?

2. Typically five-bedroom houses sell for much more than two-bedroom houses. Is

this consistent with answer (c)?

3. Lot size (Lsize) is measured in square feet. Is another scale might

be more appropriate?

Answer 1

1. 0.853> 0.05 hence we fail to reject the null hypothesis. The coefficient is not different from zero.

2. No. With an additional bedroom and keeping other variables constant, the house prices increase by 0.485 hence a 5 bedroom is more expensive than a 2 bedroom house in real world.

3. Yes. The coefficient is very small. If the lot size was measured in 1000 square feet, the coefficient would be 2 which is easier to read.