The questions below are based on a dataset containing the characteristics of 5,407
households. In particular, we will use the following variables:
- size: home size, measured in square feet.
- hhinc: annual household net income, measured in euros.
- owner: dummy variable equal to 1 if the living space is owned, and 0 if the living
space is rented.
- hhsize1: dummy variable equal to 1 if the number of household’s members is 1 or
2, and 0 otherwise.
- hhsize2: dummy variable equal to 1 if the number of household’s members is 3 or
4, and 0 otherwise.
- hhsize3: dummy variable equal to 1 if the number of household’s members is 5 or
above 5, and 0 otherwise.
- edu: education level of household’s members.
Consider the results in Table 4. Observe that the estimated coefficients of hhsize1
and hhsize2 are now negative. Does their interpretation change with respect to Table
3? Explain.
Extracted text: Table 3 . regress size hhánc owner hhsizel hhsizez hhsize3, robust noconstant Linear regression Number of obs 5,407 F(5, 5402) 11317.91 Prob > F 0.0000 R-squared 0.9157 Root MSE 348.5 Robust size Coefficient std. err. P>|t| (95% conf. interval] hhinc .005163 .0004976 18.38 0.000 .0041875 .0061384 Owner 374.5049 10.69219 35.03 0.000 353.5439 395.4659 hhsizel 647.8994 12.79973 50.62 0.000 622.8068 672.992 hhsize2 7.3956רר 21.42977 36.28 0.000 735.3846 819.4065 hhsize3 930.3489 29.34556 31.70 0.000 872.8197 987.878
Extracted text: Table 4 . regress size hhinc owner hhsizel hhsize2 hhsize3, robust Linear regression Number of obs 5,407 F(4, 5402) 877.84 Prob > F 0.0000 R-squared 0.4269 Root MSE 348.5 Robust size Coefficient std. err. P>|t| [95% conf. interval] hhinc .005163 .0004976 10.38 0.000 0041875 e061384 owner 374.5049 10.69219 35.03 0.000 353.5439 395.4659 hhsizel -282.4495 22.50563 -12.55 0.000 -326.5696 -238.3294 hhsize2 -152.9533 20.30738 -7.53 0.000 -192.764 -113.1426 hhsize3 O (omitted) cons 930.3489 29.34556 31.70 0.000 872.8197 987.878