In: Statistics and Probability
A company has recorded data on a sample of real estate listings
from Waltham, MA. The variables are:
PRICE -- List price, thousands of dollars
SQFT -- Square footage
BEDS -- Number of bedrooms
BATHS -- Number of bathrooms
HWY -- A dummy variable (1 = close to highways; 0 = far from
highways).
Use Excel's Regression tool to answer the following
questions:
PRICE |
SQFT |
BEDS |
BATHS |
HWY |
713 |
2400 |
3 |
3 |
0 |
645 |
2524 |
3 |
2 |
1 |
625 |
2732 |
4 |
2.5 |
1 |
585 |
1947 |
3 |
1.5 |
0 |
583 |
2224 |
3 |
2.5 |
0 |
540 |
1488 |
3 |
1.5 |
0 |
511 |
1752 |
3 |
1.5 |
0 |
463 |
1714 |
3 |
2 |
1 |
435 |
1500 |
3 |
1.5 |
1 |
402 |
1152 |
3 |
1 |
1 |
380 |
1272 |
3 |
1 |
1 |
368 |
1272 |
3 |
1 |
1 |
356 |
1431 |
2 |
2 |
1 |
330 |
1465 |
3 |
1 |
0 |
308 |
850 |
1 |
1 |
0 |
Fill in Multiple Blanks. For all numerical
answers, show two (2) digits to the right of the decimal point, for
example, 1.00, 1.20, 1.22. Apply the appropriate rounding rule if
necessary. Hint: You can use the “Format Cell” option in
the Regression output so that it shows two digits after the decimal
point. Excel will automatically round the values up or down, if
necessary.
1. The estimated regression line is (enter the estimated
coefficients in the appropriate space):
PRICEhat = Blank 1 + Blank 2 SQFT + Blank 3 BEDS + Blank 4 BATHS +
Blank 5 HWY
2. On average, a house with 4 bedrooms will be Blank 6 thousand
dollars Blank 7 (cheaper, more expensive) than a house with 2
bedrooms, ceteris paribus.
3. On average, a house located close to highways will be Blank 8
thousand dollars Blank 9 (cheaper, more expensive) than a house
located far from highways, ceteris paribus.
4. Predict PRICE for a house with square footage of 1960, 2
bedrooms and 2.5 bathrooms, which is located far from highways.
PRICEhat = Blank 10 (in thousands of dollars).
5. At 90% confidence, SQFT Blank 11 (is, is not) significantly
related to PRICE.
6. At 90% confidence, BEDS Blank 12 (is, is not) significantly
related to PRICE.
7. At 90% confidence, BATHS Blank 13 (is, is not) significantly
related to PRICE.
8. At 90% confidence, HWY Blank 14 (is, is not) significantly
related to PRICE.
9. True or false? At 90% confidence, a significant relationship
exists between PRICE and the set of all the independent variables
included in the regression model (SQFT, BEDS, BATHS, and HWY).
Blank 15 (true, false).
10. True or false? About 85% of the variability in PRICE is
explained by the set of all the independent variables included in
the regression model (SQFT, BEDS, BATHS, and HWY), and about 15% of
the variability in PRICE is explained by the other factors not
included in the regression. Blank 16 (true, false).
The regression output is:
R² | 0.848 | ||||||||
Adjusted R² | 0.788 | ||||||||
R | 0.921 | ||||||||
Std. Error | 58.681 | ||||||||
n | 15 | ||||||||
k | 4 | ||||||||
Dep. Var. | PRICE | ||||||||
ANOVA table | |||||||||
Source | SS | df | MS | F | p-value | ||||
Regression | 1,92,572.1786 | 4 | 48,143.0446 | 13.98 | .0004 | ||||
Residual | 34,434.7548 | 10 | 3,443.4755 | ||||||
Total | 2,27,006.9333 | 14 | |||||||
Regression output | confidence interval | ||||||||
variables | coefficients | std. error | t (df=10) | p-value | 95% lower | 95% upper | |||
Intercept | 115.21 | ||||||||
SQFT | 0.16 | 0.0712 | 2.184 | .0539 | -0.0031 | 0.3140 | |||
BEDS | 23.27 | 36.9245 | 0.630 | .5427 | -59.0021 | 105.5435 | |||
BATHS | 36.37 | 49.0941 | 0.741 | .4759 | -73.0200 | 145.7568 | |||
HWY | -49.07 | 32.1388 | -1.527 | .1578 | -120.6843 | 22.5352 | |||
Predicted values for: PRICE | |||||||||
95% Confidence Interval | 95% Prediction Interval | ||||||||
SQFT | BEDS | BATHS | HWY | Predicted | lower | upper | lower | upper | Leverage |
1,960 | 2 | 2.5 | 0 | 557.35 | 469.771 | 644.931 | 399.979 | 714.722 | 0.449 |
1. The estimated regression line is (enter the estimated coefficients in the appropriate space):
PRICEhat = 115.21 + 0.16 SQFT + 23.27 BEDS + 36.37 BATHS - 49.07 HWY
2. On average, a house with 4 bedrooms will be 46.54 thousand dollars more expensive than a house with 2 bedrooms, ceteris paribus.
3. On average, a house located close to highways will be 49.07 thousand dollars cheaper than a house located far from highways, ceteris paribus.
4. Predict PRICE for a house with square footage of 1960, 2
bedrooms and 2.5 bathrooms, which is located far from highways.
PRICEhat = Blank 10 (in thousands of dollars).
PRICEhat = 557.35 (in thousands of dollars)
5. At 90% confidence, SQFT Blank 11 is significantly related to
PRICE.
6. At 90% confidence, BEDS is not significantly related to
PRICE.
7. At 90% confidence, BATHS is not significantly related to
PRICE.
8. At 90% confidence, HWY is not significantly related to
PRICE.
9. True or false? At 90% confidence, a significant relationship
exists between PRICE and the set of all the independent variables
included in the regression model (SQFT, BEDS, BATHS, and HWY).
true
10. True or false? About 85% of the variability in PRICE is
explained by the set of all the independent variables included in
the regression model (SQFT, BEDS, BATHS, and HWY), and about 15% of
the variability in PRICE is explained by the other factors not
included in the regression. true