fit3 <- lm(review_scores_rating ~ room_type , data = airbnb)
fit3 %>% get_regression_table() %>% knitr::kable() %>% kable_styling()
term
|
estimate
|
std_error
|
statistic
|
p_value
|
lower_ci
|
upper_ci
|
intercept
|
93.202
|
0.137
|
680.602
|
0
|
92.934
|
93.471
|
room_typeHotel room
|
-4.448
|
0.667
|
-6.666
|
0
|
-5.755
|
-3.140
|
room_typePrivate room
|
0.930
|
0.197
|
4.723
|
0
|
0.544
|
1.316
|
room_typeShared room
|
-4.276
|
0.465
|
-9.203
|
0
|
-5.186
|
-3.365
|
kable(
list(
model.matrix(fit3) %>% head(),
airbnb %>% select(room_type) %>% head()
),
caption = "R's representation of categorical variables vs. what we see in the data",
valign = 't'
) %>% kable_styling()
R’s representation of categorical variables vs. what we see in the data
(Intercept)
|
room_typeHotel room
|
room_typePrivate room
|
room_typeShared room
|
1
|
0
|
1
|
0
|
1
|
0
|
0
|
0
|
1
|
0
|
1
|
0
|
1
|
0
|
1
|
0
|
1
|
0
|
0
|
0
|
1
|
0
|
0
|
0
|
|
room_type
|
Private room
|
Entire home/apt
|
Private room
|
Private room
|
Entire home/apt
|
Entire home/apt
|
|
Four cases
- \(\hat{f}(\text{entire home}) = \hat{\beta}_{0} + 0 + 0 + 0\)
- \(\hat{f}(\text{hotel}) = \hat{\beta}_{0} + \hat{\beta}_{\text{hotel}} \cdot 1 + 0 + 0\)
- \(\hat{f}(\text{private}) = \hat{\beta}_{0} + 0 + \hat{\beta}_{\text{private}} \cdot 1 + 0\)
- \(\hat{f}(\text{shared}) = \hat{\beta}_{0} + 0 + 0 + \hat{\beta}_{\text{shared}} \cdot 1\)