Variables Entered/Removedb

Model Variables Entered Variables Removed Method

dime
1 Location, Floor, . Enter
a
nsio
Price

a. All requested variables entered.

b. Dependent Variable: unitPrice

Model Summary

Model Change Statistics

Adjusted R Std. Error of the R Square

R R Square Square Estimate Change F Change df1 df2 Sig. F Change

a
dimen
1 .884 .781 .780 686.37068 .781 1133.221 3 955 .000

sion0

a. Predictors: (Constant), Location, Floor, Price

ANOVAb

Model Sum of Squares df Mean Square F Sig.

1 Regression 1.602E9 3 5.339E8 1133.221 .000a

Residual 4.499E8 955 471104.706

Total 2.052E9 958

a. Predictors: (Constant), Location, Floor, Price

b. Dependent Variable: unitPrice

Coefficientsa

Model Standardized

Unstandardized Coefficients Coefficients 95.0% Confidence Interval for B

B Std. Error Beta t Sig. Lower Bound Upper Bound

1 (Constant) 6213.222 145.485 42.707 .000 5927.714 6498.730

Price .003 .000 .440 25.160 .000 .003 .003

Floor 45.450 4.203 .168 10.815 .000 37.202 53.697

Location -624.196 21.052 -.521 -29.649 .000 -665.510 -582.881

a. Dependent Variable: unitPrice

In this case, the equation for housing price is:

2) Improved Model:

Variables Entered/Removedb

Model Variables Entered Variables Removed Method

dime
1 Floor, Q4, Price, Q1, . Enter

nsio
L, Q2, Location,

FloorQuadratic, Q3a

a. All requested variables entered.

b. Dependent Variable: unitPrice

Model Summary

Model Change Statistics

Adjusted R Std. Error of the R Square

R R Square Square Estimate Change F Change df1 df2 Sig. F Change

a
dimen
1 .892 .795 .794 664.91466 .795 410.138 9 949 .000

sion0

a. Predictors: (Constant), Floor, Q4, Price, Q1, L, Q2, Location, FloorQuadratic, Q3

ANOVAb

Model Sum of Squares df Mean Square F Sig.

1 Regression 1.632E9 9 1.813E8 410.138 .000a

Residual 4.196E8 949 442111.500

Total 2.052E9 958

a. Predictors: (Constant), Floor, Q4, Price, Q1, L, Q2, Location, FloorQuadratic, Q3

b. Dependent Variable: unitPrice

Coefficientsa

Model Unstandardized Coefficients Standardized t Sig. 95.0% Confidence Interval for B

Coefficients
 B Std. Error Beta Lower Bound Upper Bound

1 (Constant) 6310.619 730.392 8.640 .000 4877.249 7743.989

L -164.598 94.427 -.055 -1.743 .082 -349.907 20.712

Q1 882.854 675.194 .192 1.308 .191 -442.191 2207.900

Q2 736.672 675.911 .186 1.090 .276 -589.781 2063.124

Q3 365.568 675.357 .112 .541 .588 -959.799 1690.934

Q4 667.261 674.898 .227 .989 .323 -657.204 1991.726

FloorQuadratic .865 .620 .069 1.396 .163 -.352 2.082

Price .003 .000 .415 23.886 .000 .003 .003

Location -724.003 39.947 -.604 -18.124 .000 -802.397 -645.609

Floor 23.879 13.430 .088 1.778 .076 -2.478 50.235

a. Dependent Variable: unitPrice

So in the case of the improved model, we have created a dummy variable to judge whether
location is good or bad. Location was good (=1), when it was between values 2 and 4. and
location was bad (=0) when it was 5 and 6. Similarly, we created a dummy variable to judge
which quarter the house was sold, which was split up into Q1, Q2, Q3 and Q4. This meant
that when Q1 was 1, it was in the first quarter and otherwise 0. Q2 was 1 when it was in the
second quarter and otherwise 0. Q3 was 1 when it was in the third quarter and otherwise 0.
And finally Q4 was 1 when it was in the fourth quarter and otherwise 0.

In the improved model the adjusted r squared value was 0.794, and when compared to the
basic model (which was 0.780), it shows that the improved model was better since it has a
higher adjusted r squared value.

When comparing the impact of the quarters sold to the housing price, Q1 has the most impact
since when it changes, the housing price increases by 882.854 units. On the other hand, Q3
has the least impact among the quarters sold, only increasing by 365.568 units. The dummy
variable L and location are the only values that have a negative impact to the housing price
value, meaning when L changes, the housing price decreases. All other variables have
positive impacts to housing price when it increases in value.