def PlotOutliers(samples):
"""Make CDFs showing the distribution of outliers."""
cdfs = []
for label, sample in samples.iteritems():
outliers = [x for x in sample if x < 150]
cdf = thinkbayes.MakeCdfFromList(outliers, label)
cdfs.append(cdf)
thinkplot.Clf()
thinkplot.Cdfs(cdfs)
thinkplot.Save(root='variability_cdfs',
title='CDF of height',
xlabel='Reported height (cm)',
ylabel='CDF')
def PlotCdfs(d, labels):
"""Plot CDFs for each sequence in a dictionary.
Jitters the data and subtracts away the mean.
d: map from key to sequence of values
labels: map from key to string label
"""
thinkplot.Clf()
for key, xs in d.iteritems():
mu = thinkstats.Mean(xs)
xs = thinkstats.Jitter(xs, 1.3)
xs = [x - mu for x in xs]
cdf = thinkbayes.MakeCdfFromList(xs)
thinkplot.Cdf(cdf, label=labels[key])
thinkplot.Show()
def TestCorrelation(cdf):
"""Tests the correlated generator.
Makes sure that the sequence has the right distribution and correlation.
"""
n = 10000
rho = 0.4
rdt_seq = CorrelatedGenerator(cdf, rho)
xs = [rdt_seq.next() for _ in range(n)]
rho2 = correlation.SerialCorr(xs)
print(rho, rho2)
cdf2 = thinkbayes.MakeCdfFromList(xs)
thinkplot.Cdfs([cdf, cdf2])
thinkplot.Show()
def main():
ctr1 = 0.05
ctr2 = 0.05
global sample_diff
sample_diff = SampleDistOfDiff(ctr1, ctr2, n=2000)
#thinkplot.Cdf(sample_diff)
#thinkplot.Show()
sample_pval = SampleDistPval(ctr1, ctr2)
thinkplot.Cdf(sample_pval)
thinkplot.Save(root='abtest3',
xlabel='p-value',
ylabel='CDF',
formats=FORMATS)
return
sample_dist = SampleDist(ctr1, ctr2)
thinkplot.Cdf(sample_dist)
thinkplot.Save(root='abtest1',
xlabel='prob A > B',
ylabel='CDF',
formats=FORMATS)
return
sample_pred = SamplePredDist(ctr1, ctr2)
thinkplot.Cdf(sample_dist)
thinkplot.Cdf(sample_pred)
thinkplot.Save(root='abtest2',
xlabel='prob A > B',
ylabel='CDF',
formats=FORMATS)
return
# plot the prior distribution of CTR
ps = SampleCtr(100)
cdf = thinkbayes.MakeCdfFromList(ps)
thinkplot.Cdf(cdf)
thinkplot.Show()
def SamplePredDist(ctr1, ctr2, n=30):
"""Computes the sample distribution of p.
Where p is the predictive posterior probability of A>B.
ctr1: CTR of A
ctr2: CTR of B
n: number of iterations
returns: Cdf of p
"""
ps = []
for i in range(n):
pred = PredDist(ctr1, ctr2)
p = pred.Mean()
ps.append(p)
sample_pred = thinkbayes.MakeCdfFromList(ps, name='pred means')
return sample_pred
def PredDist(ctr1, ctr2, n=100):
"""Predictive posterior distribution of prob A>B.
ctr1: float CTR for A
ctr2: float CTR for B
n: number of simulations to run
returns: Cdf of posterior probs
"""
data1 = FakeData(100, ctr1)
data2 = FakeData(100, ctr2)
cdf1 = MakePosterior(data1).MakeCdf()
cdf2 = MakePosterior(data2).MakeCdf()
ctr1s = cdf1.Sample(n)
ctr2s = cdf2.Sample(n)
ps = [RunSimulation(q1, q2) for q1, q2 in zip(ctr1s, ctr2s)]
pred = thinkbayes.MakeCdfFromList(ps)
return pred
def __init__(self, prices, bids, diffs):
self.pdf_price = EstimatedPdf(prices)
self.cdf_diff = thinkbayes.MakeCdfFromList(diffs)
mu =0
sigma = numpy.std(diffs)
self.pdf_error = GaussianPdf(mu, sigma)
351.0,
286.0,
373.0,
232.0,
393.0,
745.0,
636.0,
758.0,
]
#print(OBSERVED_GAP_TIMES)
print "cumulated data number :", len(OBSERVED_GAP_TIMES)
#OBSERVED_GAP_TIMES = OBSERVED_GAP_TIMES/60 # this is not working...
for i in xrange(0, len(OBSERVED_GAP_TIMES)):
OBSERVED_GAP_TIMES[i] = OBSERVED_GAP_TIMES[i] / 60
#print(OBSERVED_GAP_TIMES)
cdf_z = thinkbayes.MakeCdfFromList(OBSERVED_GAP_TIMES)
sample_z = cdf_z.Sample(220)
pmf_z = thinkbayes.MakePmfFromList(sample_z)
#pmf_z = scipy.stats.gaussian_kde(sample_z)
thinkplot.Clf()
thinkplot.preplot(2)
thinkplot.Clf()
thinkplot.Pmf(pmf_z)
thinkplot.Save(root='chapter8_self1',
xlabel='',
ylabel='Probability',
formats=['pdf'])
