##
## Call:
## lm(formula = property_price_log ~ lot_land_area_log + living_area_ratio +
## price_per_living_area_unit_usd, data = zillowmorgantown)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.7531 -0.0795 0.1185 0.2926 0.5947
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) 10.7195287 0.3290716 32.575
## lot_land_area_log -0.1169859 0.0317447 -3.685
## living_area_ratio 0.0411534 0.0036838 11.171
## price_per_living_area_unit_usd 0.0122438 0.0005841 20.962
## Pr(>|t|)
## (Intercept) < 0.0000000000000002 ***
## lot_land_area_log 0.000274 ***
## living_area_ratio < 0.0000000000000002 ***
## price_per_living_area_unit_usd < 0.0000000000000002 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5628 on 280 degrees of freedom
## Multiple R-squared: 0.6596, Adjusted R-squared: 0.656
## F-statistic: 180.9 on 3 and 280 DF, p-value: < 0.00000000000000022
##
## Call:
## lm(formula = property_price_usd ~ living_area_ratio + price_per_living_area_unit_usd,
## data = zillowmorgantown)
##
## Residuals:
## Min 1Q Median 3Q Max
## -625107 -19653 -2110 27792 113073
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -111893.3 11850.6 -9.442 <0.0000000000000002
## living_area_ratio 7538.5 373.6 20.178 <0.0000000000000002
## price_per_living_area_unit_usd 1424.1 59.6 23.893 <0.0000000000000002
##
## (Intercept) ***
## living_area_ratio ***
## price_per_living_area_unit_usd ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 57650 on 281 degrees of freedom
## Multiple R-squared: 0.7507, Adjusted R-squared: 0.7489
## F-statistic: 423 on 2 and 281 DF, p-value: < 0.00000000000000022
##
## Call:
## lm(formula = property_price_usd ~ price_per_living_area_unit_usd +
## bedrooms + bathrooms, data = zillowmorgantown)
##
## Residuals:
## Min 1Q Median 3Q Max
## -400396 -35251 -210 36334 188033
##
## Coefficients:
## Estimate Std. Error t value
## (Intercept) -106933.92 13673.78 -7.820
## price_per_living_area_unit_usd 1278.08 65.58 19.490
## bedrooms 28840.72 3694.59 7.806
## bathrooms 31951.03 4324.36 7.389
## Pr(>|t|)
## (Intercept) 0.000000000000111 ***
## price_per_living_area_unit_usd < 0.0000000000000002 ***
## bedrooms 0.000000000000122 ***
## bathrooms 0.000000000001755 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 63300 on 277 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.6972, Adjusted R-squared: 0.6939
## F-statistic: 212.6 on 3 and 277 DF, p-value: < 0.00000000000000022
## lot_land_area_sqft property_price_usd bedrooms
## lot_land_area_sqft 1.0000000 -0.2248229 -0.2551373
## property_price_usd -0.2248229 1.0000000 0.4583067
## bedrooms -0.2551373 0.4583067 1.0000000
```
Both M and m2 yield better r^2 values than m3 although I think m3 is the best determinent for the cost of the house, despite what the data says. There are many ways to get to a high r^2 with this data, but I have not found one that yielded a logical result in terms of a positive price for a house as the intercept.