def MakeFigures(firsts, others):
"""Plot Hists and Pmfs for pregnancy length.
firsts: DataFrame
others: DataFrame
"""
first_hist = thinkstats2.MakeHistFromList(firsts.prglngth)
first_pmf = thinkstats2.MakePmfFromHist(first_hist)
other_hist = thinkstats2.MakeHistFromList(others.prglngth)
other_pmf = thinkstats2.MakePmfFromHist(other_hist)
width = 0.4
first_options = dict(label='first', width=-width)
other_options = dict(label='other', width=width)
# plot the histograms
thinkplot.PrePlot(2)
thinkplot.Hist(first_hist, **first_options)
thinkplot.Hist(other_hist, **other_options)
axis = [27, 46, 0, 2700]
thinkplot.Save(root='nsfg_hist',
title='Histogram',
xlabel='weeks',
ylabel='frequency',
axis=axis)
# plot the PMFs
thinkplot.PrePlot(2)
thinkplot.Hist(first_pmf, **first_options)
thinkplot.Hist(other_pmf, **other_options)
axis = [27, 46, 0, 0.6]
thinkplot.Save(root='nsfg_pmf',
title='PMF',
xlabel='weeks',
ylabel='probability',
axis=axis)
# plot the differences in the PMFs
weeks = range(35, 46)
diffs = []
for week in weeks:
p1 = first_pmf.Prob(week)
p2 = other_pmf.Prob(week)
diff = 100 * (p1 - p2)
diffs.append(diff)
thinkplot.PrePlot(1)
thinkplot.Bar(weeks, diffs, align='center')
thinkplot.Save(root='nsfg_diffs',
title='Difference in PMFs',
xlabel='weeks',
ylabel='percentage points',
legend=False)
def pmf_diff_in_percentage(first_pmf, other_pmf):
"""This figure makes the pattern clearer: first babies are less likely to be born
in week 39, and somewhat more likely to be born in weeks 41 and 42"""
weeks = range(35, 46)
# To look near mode
diffs = []
for week in weeks:
p1 = first_pmf.Prob(week)
p2 = other_pmf.Prob(week)
diff = 100 * (p1 - p2)
diffs.append(diff)
thinkplot.Bar(weeks, diffs)
thinkplot.Show(xlabel="Weeks", ylabel="pmf % diff")
def MakeFigures(firsts, others):
"""Plot Pmfs of pregnancy length.
firsts: DataFrame
others: DataFrame
"""
# plot the PMFs
first_pmf = thinkstats2.Pmf(firsts.prglngth, label='first')
other_pmf = thinkstats2.Pmf(others.prglngth, label='other')
width = 0.45
thinkplot.PrePlot(2, cols=2)
thinkplot.Hist(first_pmf, align='right', width=width)
thinkplot.Hist(other_pmf, align='left', width=width)
thinkplot.Config(xlabel='weeks',
ylabel='probability',
axis=[27, 46, 0, 0.6])
thinkplot.PrePlot(2)
thinkplot.SubPlot(2)
thinkplot.Pmfs([first_pmf, other_pmf])
thinkplot.Save(root='probability_nsfg_pmf',
xlabel='weeks',
axis=[27, 46, 0, 0.6])
# plot the differences in the PMFs
weeks = range(35, 46)
diffs = []
for week in weeks:
p1 = first_pmf.Prob(week)
p2 = other_pmf.Prob(week)
diff = 100 * (p1 - p2)
diffs.append(diff)
thinkplot.Bar(weeks, diffs)
thinkplot.Save(root='probability_nsfg_diffs',
title='Difference in PMFs',
xlabel='weeks',
ylabel='percentage points',
legend=False)
thinkplot.PrePlot(2)
thinkplot.Pmfs([first_pmf, others_pmf])
thinkplot.show(xlabl='week', ylabel='probability', axis=[27, 46, 0, 0.6])
# %% [markdown]
# ## 3.3 その他の可視化
# %%
# 差を棒グラフで表示
weeks = range(35, 46)
diffs = []
for week in weeks:
p1 = first_pmf.Prob(week)
p2 = others_pmf.Prob(week)
diff = 100 * (p1 - p2)
diffs.append(diff)
thinkplot.Bar(weeks, diffs)
# %% [markdown]
# ## 3.4 クラスサイズのパラドックス
# %%
d = {7: 8, 12: 8, 17: 14, 22: 4, 27: 6, 32: 12, 37: 8, 42: 3, 47: 2}
pmf = thinkstats2.Pmf(d, label='actual')
print('mean of the actual pmf:', pmf.Mean())
# %%
def BiasPmf(pmf, label):
"""バイアスされたPmfを推定する
Arguments:
pmf {Pmf} -- actual pmf
label {string} -- label
Returns: