def main():
results = ReadResults()
print '# of samples:', len(results)
results = CondFilter(results)
print '# of samples(age=30s):', len(results)
h_speed, h_rank, h_rank_age = FindProfHeo(results)
speeds = GetSpeeds(results)
pmf = Pmf.MakePmfFromList(speeds, 'speeds')
cdf = Cdf.MakeCdfFromList(speeds, 'speeds')
if h_speed:
print '허준영s speed is ', h_speed
print 'His percentile rank is ', cdf.Prob(h_speed) * 100
myplot.Clf()
myplot.Pmf(pmf)
myplot.Save(root='mbc_marathon_pmf',
title='PMF of running speed',
xlabel='speed (kmh)',
ylabel='probability')
myplot.Clf()
myplot.Cdf(cdf)
myplot.Save(root='mbc_marathon_cdf',
title='CDF of running speed',
xlabel='speed (kmh)',
ylabel='probability')
myplot.Close()
def MakeFigures(pool):
myplot.Clf()
myplot.Hist(pool.weight_pmf, linewidth=1, color='blue')
myplot.Save(root='nsfg_birthwgt_pmf',
title='Birth weight PMF',
xlabel='weight (ounces)',
ylabel='probability')
myplot.Clf()
myplot.Cdf(pool.lengths_pmf, linewidth=1, color='blue')
myplot.Save(root='nsfg_weeks_pmf',
title='Gestation PMF',
xlabel='weeks',
ylabel='probability')
def ex6_14(n, do_log=False, ns=1000):
exp = randvar.Exponential(1)
def prod_log(t):
p = 1
for x in t:
p = p * x
if do_log == True:
return math.log(p)
else:
return p
s = [prod_log([exp.generate() for x in range(n)]) for i in range(ns)]
mu, var = thinkstats.MeanVar(s)
print n, mu, var
cdf = Cdf.MakeCdfFromList(s)
myplot.Clf()
myplot.Cdf(cdf)
myplot.Save('clt14' + str(do_log) + str(n))
rankit.MakeNormalPlot(s, 'clt14npp' + str(do_log) + str(n))
myplot.Close()
def PlotCdf(root, cdf, delta):
"""Draws a Cdf with vertical lines at the observed delta.
Args:
root: string used to generate filenames
cdf: Cdf object
delta: float observed difference in means
"""
def VertLine(x):
"""Draws a vertical line at x."""
xs = [x, x]
ys = [0, 1]
pyplot.plot(xs, ys, linewidth=2, color='0.7')
VertLine(-delta)
VertLine(delta)
xs, ys = cdf.Render()
pyplot.subplots_adjust(bottom=0.11)
pyplot.plot(xs, ys, linewidth=2, color='blue')
myplot.Save(root,
title='Resampled differences',
xlabel='difference in means',
ylabel='CDF(x)',
legend=False)
myplot.Close()
def ex6_12(n, ns=1000):
exp = randvar.Exponential(1)
s = [sum([exp.generate() for x in range(n)]) / float(n) for i in range(ns)]
mu, var = thinkstats.MeanVar(s)
print n, mu, var
cdf = Cdf.MakeCdfFromList(s)
myplot.Clf()
myplot.Cdf(cdf)
myplot.Save('clt' + str(n))
myplot.Close()
def ex6_13(n, ns=1000):
exp = randvar.Exponential(1)
s = [sum([exp.generate() for x in range(n)]) for i in range(ns)]
mu, var = thinkstats.MeanVar(s)
print n, mu, var
cdf = Cdf.MakeCdfFromList(s)
myplot.Clf()
myplot.Cdf(cdf)
myplot.Save('clt13' + str(n))
rankit.MakeNormalPlot(s, 'clt13npp' + str(n))
myplot.Close()
26 ,
23 ,
14 ,
101 ,
125 ,
0 ,
53 ,
10 ,
56 ,
165 ,
46 ,
5 ,
0 ,
38 ,
29 ,
4 ,
30 ,
89 ,
14 ,
]
cdf = Cdf.MakeCdfFromList(birthday_arrival)
print 'Mean:', cdf.Mean()
myplot.Clf()
myplot.Cdf(cdf)
myplot.Save('birthday_cdf')
myplot.Clf()
myplot.Cdf(cdf, complement=True)
myplot.Save('birthday_ccdf', yscale='log')
myplot.Close()
from thinkstats import rankit
from thinkstats import myplot
import random
sample = [random.normalvariate(0.0, 1.0) for i in range(100)]
means = rankit.EstimateRankits(len(sample))
myplot.Plot(means, sorted(sample), 'b.', markersize=3)
myplot.Save('rankit_npp', xlabel='rankits')
rankit.MakeNormalPlot(sample, 'dirty_npp')
myplot.Close()