def Process(table, name):
descriptive.Process(table, name)
table.weights = [
p.totalwgt_oz for p in table.records if p.totalwgt_oz != 'NA'
]
table.weight_pmf = Pmf.MakePmfFromList(table.weights, table.name)
table.weight_cdf = Cdf.MakeCdfFromList(table.weights, table.name)
table.lengths_pmf = Pmf.MakePmfFromList(table.lengths, table.name)
table.legnths_cdf = Cdf.MakeCdfFromList(table.lengths, table.name)
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 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 PValue(model1, model2, n, m, delta, iters=1000):
deltas = [Resample(model1, model2, n, m) for i in range(iters)]
cdf = Cdf.MakeCdfFromList(deltas)
# compute the two tail probabilities
left = cdf.Prob(-delta)
right = 1.0 - cdf.Prob(delta)
pvalue = left + right
return cdf, pvalue
def poincare_bread(n=4, nsamples=365):
poincare = [
max([random.normalvariate(950, 50) for x in range(n)])
for y in range(nsamples)
]
mu, var = ts.MeanVar(poincare)
sd = math.sqrt(var)
cdf = Cdf.MakeCdfFromList(poincare)
cdf.name = 'poincare'
return cdf, mu, sd
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 Cdf from thinkstats import myplot import random samples = [random.expovariate(2) for x in range(80)] cdf = Cdf.MakeCdfFromList(samples) myplot.Clf() myplot.Cdf(cdf, complement=True) myplot.Show(yscale='log')
from thinkstats import Cdf from thinkstats import myplot t = [2, 1, 3, 2, 5] cdf = Cdf.MakeCdfFromList(t) cdf.name = 'Sample' print 'cdf.Prob(2):', cdf.Prob(2) print 'cdf.Prob(3):', cdf.Prob(3) print 'cdf.Prob(6):', cdf.Prob(6) print 'cdf.Value(0.5):', cdf.Value(0.5) print 'cdf.Value(0.7):', cdf.Value(0.7) myplot.Clf() myplot.Cdf(cdf) myplot.Show(title='CDF', xlabel='x', ylabel='CDF(x)', axis=[0, 6, 0, 1])
from thinkstats import Cdf
from thinkstats import myplot
word_count = {}
fp = open('thinkstats.txt', 'rt')
text = fp.read()
words = text.split()
print len(words)
for word in words:
word_count[word] = word_count.get(word, 0) + 1
cdf = Cdf.MakeCdfFromList(word_count.values())
print 'Median:', cdf.Value(0.5)
print 'Mean:', cdf.Mean()
print 'Mode:', cdf.Value(1.0)
myplot.Clf()
myplot.Cdf(cdf, complement=True)
myplot.Show(xscale='log', yscale='log')
from thinkstats import myplot
def poincare_bread(n=4, nsamples=365):
poincare = [
max([random.normalvariate(950, 50) for x in range(n)])
for y in range(nsamples)
]
mu, var = ts.MeanVar(poincare)
sd = math.sqrt(var)
cdf = Cdf.MakeCdfFromList(poincare)
cdf.name = 'poincare'
return cdf, mu, sd
if __name__ == '__main__':
nsamples = 365
cdf_p, mu, sd = poincare_bread(4, nsamples)
print mu, sd
normal = [random.normalvariate(mu, sd) for y in range(nsamples)]
cdf_n = Cdf.MakeCdfFromList(normal)
cdf_n.name = 'normal'
myplot.Clf()
myplot.Cdfs([cdf_p, cdf_n])
myplot.Show()
myplot.Close()