8.32 Lab: Tree-Based Methods - Fitting Classification Trees

We will use the Carseats data set from the ISLR package to demonstrate a classification model.

## Loading required package: rpart

## 
## Attaching package: 'rpart'

## The following object is masked from 'package:dials':
## 
##     prune

head(Carseats)

##   Sales CompPrice Income Advertising Population Price ShelveLoc Age Education
## 1  9.50       138     73          11        276   120       Bad  42        17
## 2 11.22       111     48          16        260    83      Good  65        10
## 3 10.06       113     35          10        269    80    Medium  59        12
## 4  7.40       117    100           4        466    97    Medium  55        14
## 5  4.15       141     64           3        340   128       Bad  38        13
## 6 10.81       124    113          13        501    72       Bad  78        16
##   Urban  US
## 1   Yes Yes
## 2   Yes Yes
## 3   Yes Yes
## 4   Yes Yes
## 5   Yes  No
## 6    No Yes

skimr::skim(Carseats)

(#tab:skim_dataset)Data summary

Name	Carseats
Number of rows	400
Number of columns	11
_______________________
Column type frequency:
factor	3
numeric	8
________________________
Group variables	None

Variable type: factor

skim_variable	n_missing	complete_rate	ordered	n_unique	top_counts
ShelveLoc	0	1	FALSE	3	Med: 219, Bad: 96, Goo: 85
Urban	0	1	FALSE	2	Yes: 282, No: 118
US	0	1	FALSE	2	Yes: 258, No: 142

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
Sales	0	1	7.50	2.82	0	5.39	7.49	9.32	16.27	▁▆▇▃▁
CompPrice	0	1	124.97	15.33	77	115.00	125.00	135.00	175.00	▁▅▇▃▁
Income	0	1	68.66	27.99	21	42.75	69.00	91.00	120.00	▇▆▇▆▅
Advertising	0	1	6.64	6.65	0	0.00	5.00	12.00	29.00	▇▃▃▁▁
Population	0	1	264.84	147.38	10	139.00	272.00	398.50	509.00	▇▇▇▇▇
Price	0	1	115.80	23.68	24	100.00	117.00	131.00	191.00	▁▂▇▆▁
Age	0	1	53.32	16.20	25	39.75	54.50	66.00	80.00	▇▆▇▇▇
Education	0	1	13.90	2.62	10	12.00	14.00	16.00	18.00	▇▇▃▇▇

We create a new variable High to denote if Sales <= 8, then the Sales predictor is removed as it is a perfect predictor of High.

carseats <- as_tibble(Carseats) %>%
  mutate(High = factor(if_else(Sales > 8, "Yes", "No"))) %>%
  select(-Sales)