250
200
150
100
Total
50
0
1806
1852
1866
1879
1887
1894
1901
1908
1915
1922
1929
1936
1943
1950
1957
1964
1971
1978
1985
1992
1999
2006
2013
Number of Homes Sold
Total
Year Homes Sold
1.
2. a. The R-squared value of 0.3289 means that this model only explains
approximately 32.9% of the variation in Sale_amount. There are other factors not
in this data set that explain the other 67.1%.
b. The coefficients represent the predicted change in the Sale_amount for every
one-unit increase in that specific variable, assuming all other variables remain
constant.
1) Beds (-256.69): Each additional bedroom is associated with a decrease of
about $257 in sale price (though this is not statistically reliable according
to the p-values).
2) Baths (123,906.10): Adding one bathroom increases the predicted price by
$123,906. This is the data set’s most influential feature.
3) Sqft_home (51.53): Each additional square foot of living space adds about
$51.53 to the price.
4) Sqft_lot (0.21): Each additional square foot of lot size adds about $0.21 to
the price.
5) Type_Single (67,147.22): Being a Single Family home increases the
predicted value by $67,147 compared to the baseline property type.
6) Type_Multi (47,414.95): Being a Multi-Family home increases the
predicted value by $47,415 compared to the baseline.
7) Build_year (-2,002.46): For every year newer a house is, the price actually
decreases by about $2,002. (This suggests older homes in this dataset
are more valuable).
c. 1) Significant Predictors:
a) Baths, Sqft_home, Sqft_lot, Type_Single, and Build_year all have very low
p-values. You can be highly confident these factors truly impact the sale
price.
2)Non-Significant Predictors:
a) Beds (0.941): This is very high. It means that once you know the square
footage and number of bathrooms, the number of bedrooms doesn't
actually help predict the price.
b) Type_Multi (0.190): This is above 0.05, meaning there isn't enough
evidence to say that being a Multi-Family home significantly changes the
price compared to your baseline.
3. I would advise the investor to lean toward Single Family homes for a more
predictable valuation premium. For Multi-Family opportunities, they should ignore
the "type" premium and instead evaluate the deal based on the number
of bathrooms and total living square footage, as these are the true drivers of sale
price in this market.
4. It tells us that we can be 95% confident that the true effect of a variable falls
somewhere between those two numbers. The narrower the gap between the
Lower and Upper values, the more precise our model's prediction is.