1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
|
#!/usr/bin/env python
# DO NOT EDIT
# Autogenerated from the notebook plots_boxplots.ipynb.
# Edit the notebook and then sync the output with this file.
#
# flake8: noqa
# DO NOT EDIT
# # Box Plots
# The following illustrates some options for the boxplot in statsmodels.
# These include `violin_plot` and `bean_plot`.
import numpy as np
import matplotlib.pyplot as plt
import statsmodels.api as sm
# ## Bean Plots
# The following example is taken from the docstring of `beanplot`.
#
# We use the American National Election Survey 1996 dataset, which has
# Party
# Identification of respondents as independent variable and (among other
# data) age as dependent variable.
data = sm.datasets.anes96.load_pandas()
party_ID = np.arange(7)
labels = [
"Strong Democrat",
"Weak Democrat",
"Independent-Democrat",
"Independent-Independent",
"Independent-Republican",
"Weak Republican",
"Strong Republican",
]
# Group age by party ID, and create a violin plot with it:
plt.rcParams["figure.subplot.bottom"] = 0.23 # keep labels visible
plt.rcParams["figure.figsize"] = (10.0, 8.0) # make plot larger in notebook
age = [data.exog["age"][data.endog == id] for id in party_ID]
fig = plt.figure()
ax = fig.add_subplot(111)
plot_opts = {
"cutoff_val": 5,
"cutoff_type": "abs",
"label_fontsize": "small",
"label_rotation": 30,
}
sm.graphics.beanplot(age, ax=ax, labels=labels, plot_opts=plot_opts)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
# plt.show()
def beanplot(data, plot_opts={}, jitter=False):
"""helper function to try out different plot options"""
fig = plt.figure()
ax = fig.add_subplot(111)
plot_opts_ = {
"cutoff_val": 5,
"cutoff_type": "abs",
"label_fontsize": "small",
"label_rotation": 30,
}
plot_opts_.update(plot_opts)
sm.graphics.beanplot(data,
ax=ax,
labels=labels,
jitter=jitter,
plot_opts=plot_opts_)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
fig = beanplot(age, jitter=True)
fig = beanplot(age, plot_opts={"violin_width": 0.5, "violin_fc": "#66c2a5"})
fig = beanplot(age, plot_opts={"violin_fc": "#66c2a5"})
fig = beanplot(age,
plot_opts={
"bean_size": 0.2,
"violin_width": 0.75,
"violin_fc": "#66c2a5"
})
fig = beanplot(age, jitter=True, plot_opts={"violin_fc": "#66c2a5"})
fig = beanplot(age,
jitter=True,
plot_opts={
"violin_width": 0.5,
"violin_fc": "#66c2a5"
})
# ## Advanced Box Plots
# Based of example script `example_enhanced_boxplots.py` (by Ralf Gommers)
import numpy as np
import matplotlib.pyplot as plt
import statsmodels.api as sm
# Necessary to make horizontal axis labels fit
plt.rcParams["figure.subplot.bottom"] = 0.23
data = sm.datasets.anes96.load_pandas()
party_ID = np.arange(7)
labels = [
"Strong Democrat",
"Weak Democrat",
"Independent-Democrat",
"Independent-Independent",
"Independent-Republican",
"Weak Republican",
"Strong Republican",
]
# Group age by party ID.
age = [data.exog["age"][data.endog == id] for id in party_ID]
# Create a violin plot.
fig = plt.figure()
ax = fig.add_subplot(111)
sm.graphics.violinplot(
age,
ax=ax,
labels=labels,
plot_opts={
"cutoff_val": 5,
"cutoff_type": "abs",
"label_fontsize": "small",
"label_rotation": 30,
},
)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
ax.set_title("US national election '96 - Age & Party Identification")
# Create a bean plot.
fig2 = plt.figure()
ax = fig2.add_subplot(111)
sm.graphics.beanplot(
age,
ax=ax,
labels=labels,
plot_opts={
"cutoff_val": 5,
"cutoff_type": "abs",
"label_fontsize": "small",
"label_rotation": 30,
},
)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
ax.set_title("US national election '96 - Age & Party Identification")
# Create a jitter plot.
fig3 = plt.figure()
ax = fig3.add_subplot(111)
plot_opts = {
"cutoff_val": 5,
"cutoff_type": "abs",
"label_fontsize": "small",
"label_rotation": 30,
"violin_fc": (0.8, 0.8, 0.8),
"jitter_marker": ".",
"jitter_marker_size": 3,
"bean_color": "#FF6F00",
"bean_mean_color": "#009D91",
}
sm.graphics.beanplot(age,
ax=ax,
labels=labels,
jitter=True,
plot_opts=plot_opts)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
ax.set_title("US national election '96 - Age & Party Identification")
# Create an asymmetrical jitter plot.
ix = data.exog["income"] < 16 # incomes < $30k
age = data.exog["age"][ix]
endog = data.endog[ix]
age_lower_income = [age[endog == id] for id in party_ID]
ix = data.exog["income"] >= 20 # incomes > $50k
age = data.exog["age"][ix]
endog = data.endog[ix]
age_higher_income = [age[endog == id] for id in party_ID]
fig = plt.figure()
ax = fig.add_subplot(111)
plot_opts["violin_fc"] = (0.5, 0.5, 0.5)
plot_opts["bean_show_mean"] = False
plot_opts["bean_show_median"] = False
plot_opts["bean_legend_text"] = r"Income < \$30k"
plot_opts["cutoff_val"] = 10
sm.graphics.beanplot(
age_lower_income,
ax=ax,
labels=labels,
side="left",
jitter=True,
plot_opts=plot_opts,
)
plot_opts["violin_fc"] = (0.7, 0.7, 0.7)
plot_opts["bean_color"] = "#009D91"
plot_opts["bean_legend_text"] = r"Income > \$50k"
sm.graphics.beanplot(
age_higher_income,
ax=ax,
labels=labels,
side="right",
jitter=True,
plot_opts=plot_opts,
)
ax.set_xlabel("Party identification of respondent.")
ax.set_ylabel("Age")
ax.set_title("US national election '96 - Age & Party Identification")
# Show all plots.
# plt.show()
|
