Manually predicting house prices
You can manually calculate the predictions from the model coefficients. When making predictions in real life, it is better to use predict()
, but doing this manually is helpful to reassure yourself that predictions aren't magic—they are simply arithmetic.
In fact, for a simple linear regression, the predicted value is just the intercept plus the slope times the explanatory variable.
$$response = intercept + slope * explanatory$$
mdl_price_vs_conv
and explanatory_data
are available, and dplyr
is loaded.
This exercise is part of the course
Introduction to Regression in R
Exercise instructions
- Get the coefficients of
mdl_price_vs_conv
, assigning tocoeffs
. - Get the intercept, which is the first element of
coeffs
, assigning tointercept
. - Get the slope, which is the second element of
coeffs
, assigning toslope
. - Manually predict
price_twd_msq
using the intercept, slope, andn_convenience
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Get the coefficients of mdl_price_vs_conv
coeffs <- ___
# Get the intercept
intercept <- ___
# Get the slope
slope <- ___
explanatory_data %>%
mutate(
# Manually calculate the predictions
price_twd_msq = ___
)
# Compare to the results from predict()
predict(mdl_price_vs_conv, explanatory_data)