def print_critics_users_diff(users_scores, critics_scores):
users_scores_mean = users_scores.mean()
critics_scores_mean = critics_scores.mean()
print("Mean CRITIC score:", "%.2f" % critics_scores_mean)
print("Mean USER score:", "%.2f" % users_scores_mean)
print("\nUSER mean - CRITIC mean:",
"%.2f" % (users_scores_mean - critics_scores_mean))
print(
"Users rated albums",
"%.2f" % ((users_scores_mean - critics_scores_mean) * 100 /
critics_scores_mean), "percent higher than critics did.")
print("\nEffect size (Cohen's D) of",
"%.2f" % ts2.CohenEffectSize(users_scores, critics_scores))
types = ("Users", "Critics")
ypos = np.arange(len(types))
scores = [users_scores_mean, critics_scores_mean]
plt.bar(ypos, scores, align='center', alpha=0.5)
plt.xticks(ypos, types)
plt.ylabel('Mean score')
plt.title('Mean album review score (out of 100)')
plt.ylim(60, 90)
plt.show()
def main(script):
"""Tests the functions in this module.
script: string script name
"""
live, firsts, others = first.MakeFrames()
hist = thinkstats2.Hist(live.prglngth)
# test Mode
mode = Mode(hist)
print('Mode of preg length', mode)
assert mode == 39, mode
# test AllModes
modes = AllModes(hist)
assert modes[0][1] == 4693, modes[0][1]
for value, freq in modes[:5]:
print(value, freq)
firsts_wgt = firsts.totalwgt_lb.mean()
others_wgt = others.totalwgt_lb.mean()
print("firsts = {} pounds, others = {} pounds, dif = {} pounds ".format(
firsts_wgt, others_wgt, firsts_wgt - others_wgt))
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print(d)
print('%s: All tests passed.' % script)
def Summarize(live, firsts, others):
"""Print various summary statistics."""
mean = live.prglngth.mean()
var = live.prglngth.var()
std = live.prglngth.std()
print('Live mean', mean)
print('Live variance', var)
print('Live std', std)
mean1 = firsts.prglngth.mean()
mean2 = others.prglngth.mean()
var1 = firsts.prglngth.var()
var2 = others.prglngth.var()
print('Mean')
print('First babies', mean1)
print('Others', mean2)
print('Variance')
print('First babies', var1)
print('Others', var2)
print('Difference in weeks', mean1 - mean2)
print('Difference in hours', (mean1 - mean2) * 7 * 24)
print('Difference relative to 39 weeks', (mean1 - mean2) / 39 * 100)
d = thinkstats2.CohenEffectSize(firsts.prglngth, others.prglngth)
print('Cohen d', d)
def WeightDifference(live, firsts, others):
"""Explore the difference in weight between first babies and others.
live: DataFrame of all live births
firsts: DataFrame of first babies
others: DataFrame of others
"""
mean0 = live.totalwgt_lb.mean()
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
var1 = firsts.totalwgt_lb.var()
var2 = others.totalwgt_lb.var()
print('Mean')
print('First babies', mean1)
print('Others', mean2)
print('Variance')
print('First babies', var1)
print('Others', var2)
print('Difference in lbs', mean1 - mean2)
print('Difference in oz', (mean1 - mean2) * 16)
print('Difference relative to mean (%age points)',
(mean1 - mean2) / mean0 * 100)
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen d', d)
def WeightDifference(live, firsts, others):
# calculate mean for difference in lbs
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
print('Difference in lbs', mean1 - mean2) # firsts are lighter than others
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen d', d)
return d
def wt_diff(live, firsts, others):
mean0 = live.totalwgt_lb.mean()
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
var1 = firsts.totalwgt_lb.var()
var2 = others.totalwgt_lb.var()
print("mean: ", var1, var2)
print("diff_lb: ", mean1 - mean2)
print("diff_rel_mean: ", (mean1 - mean2) / mean0 * 100)
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print("cohen's d: ", d)
def WeightComparison():
preg = nsfg.ReadFemPreg()
live = preg[preg.outcome == 1]
firsts = live[live.birthord == 1]
others = live[live.birthord != 1]
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print(d)
print(
"Still a really small effect size, though not as small as difference in pregnancy length"
)
def WeightDifferences(first, other, live):
print("Means:")
first_mean = first.totalwgt_lb.mean()
print("First mean:", first_mean)
other_mean = other.totalwgt_lb.mean()
print("Other mean:", other_mean)
live_mean = live.totalwgt_lb.mean()
print("Live mean:", live_mean)
print("Means differences:", (abs(first_mean - other_mean) / live_mean)*100)
d = thinkstats2.CohenEffectSize(first.totalwgt_lb, other.totalwgt_lb)
return d
def WeightDiff(firsts, others):
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
var1 = firsts.totalwgt_lb.var()
var2 = others.totalwgt_lb.var()
print('Mean')
print('First babies ', mean1)
print('Others ', mean2)
print('Variance')
print('First babies ', var1)
print('Others ', var2)
print('Difference mean that First babies and others ', mean1 - mean2)
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen d ', d)
def Summarize(greq, less):
"""Print various summary statistics."""
gMean = greq.prglngth.mean()
gVar = greq.prglngth.var()
print '>= 30 mean:', gMean
print '>= 30 variance:', gVar
lMean = less.prglngth.mean()
lVar = less.prglngth.var()
print '< 30 mean:', lMean
print '< 30 variance:', lVar
print 'Difference between means:', (abs(gMean - lMean))
d = thinkstats2.CohenEffectSize(greq.prglngth, less.prglngth)
print 'Cohen d:', d
def main(script):
"""Tests the functions in this module.
script: string script name
"""
live, firsts, others = first.MakeFrames()
hist = thinkstats2.Hist(live.prglngth)
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
var1 = firsts.totalwgt_lb.var()
var2 = others.totalwgt_lb.var()
print('Mean Weight')
print('First babies', mean1)
print('Others babies', mean2)
print('Variance in Weight')
print('First babies', var1)
print('Others babies', var2)
print('Difference in lbs', mean1 - mean2)
cohen_d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb,
others.totalwgt_lb)
print('Cohen d', cohen_d)
# test Mode
mode = Mode(hist)
print('Mode of preg length', mode)
assert mode == 39, mode
# test AllModes
modes = AllModes(hist)
assert modes[0][1] == 4693, modes[0][1]
for value, freq in modes[:5]:
print(value, freq)
print('%s: All tests passed.' % script)
def CompareWeight(live, firsts, others):
meanlive = live.totalwgt_lb.mean()
meanfirsts = firsts.totalwgt_lb.mean()
meanothers = others.totalwgt_lb.mean()
print('Mean')
print('Firsts', meanfirsts)
print('Others', meanothers)
print('Difference in lbs', meanfirsts-meanothers)
print('Difference in %', (meanfirsts-meanothers) / meanlive * 100)
varfirsts = firsts.totalwgt_lb.var()
varothers = others.totalwgt_lb.var()
print('Variance')
print('Firsts', varfirsts)
print('Others', varothers)
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen effect', d)
def light_heavy(firsts, others):
print(firsts.totalwgt_lb.mean(), others.totalwgt_lb.mean())
print(thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb))
return ""
import math, thinkstats2, nsfg
preg = nsfg.ReadFemPreg()
live = preg[preg.outcome == 1]
firsts = live[live.birthord == 1]
others = live[live.birthord != 1]
mean_f = firsts.totalwgt_lb.mean()
mean_o = others.totalwgt_lb.mean()
var_f = firsts.totalwgt_lb.var()
var_o = others.totalwgt_lb.var()
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('First babies average weight', mean_f)
print('Other babies average weight', mean_o)
print('Difference in weight between first and other babies, in oz:',
(mean_f - mean_o) * 16)
print('Difference in weight between first and other babies, % :',
(mean_f - mean_o) * 100 / mean_o)
print('First babies variance', var_f)
print('Other babies variance', var_o)
print('Cohen d for weight', d)
n_f = len(firsts)
n_o = len(others)
self.n, self.m = len(group1), len(group2)
self.pool = np.hstack((group1, group2))
def RunModel(self):
np.random.shuffle(self.pool)
data = self.pool[:self.n], self.pool[self.n:]
return data
male, female = MakeFrames()
alcwknd = male.alcwknd, female.alcwknd
MakeHists(male, female)
MakePmfs(male, female)
MakeStep(male, female)
MakeCdfs(male, female)
ht = DiffMeansPermute(alcwknd)
pvalue = ht.PValue()
ht.PlotCdf()
thinkplot.Config(xlabel='Difference in Means',
ylabel='CDF',
title='Weekend Alcohol Consumption')
thinkplot.Show()
print 'Weekend Alcohol Consumption:\nMen:\nMean:', male.alcwknd.mean(
), '\nVariance:', male.alcwknd.var()
print '\nWomen:\nMean:', female.alcwknd.mean(
), '\nVariance:', female.alcwknd.var()
print '\nDifference in means:', abs(male.alcwknd.mean() -
female.alcwknd.mean())
print '\nCohen d:', thinkstats2.CohenEffectSize(male.alcwknd, female.alcwknd)
print '\np-value:', pvalue
print('Others', var2)
print('Difference in lbs', mean1 - mean2)
print('Difference in oz', (mean1 - mean2) * 16)
print('Difference relative to mean (%age points)',
(mean1 - mean2) / mean0 * 100)
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen d', d)
# In[319]:
d = thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb)
print('Cohen d', d)
# In[325]:
mean0 = live.totalwgt_lb.mean()
mean1 = firsts.totalwgt_lb.mean()
mean2 = others.totalwgt_lb.mean()
print('Mean')
print('First babies', mean1)
print('Others', mean2)
group2 = np.random.choice(self.pool, self.m, replace=True)
data = group1, group2
return data
preg = nsfg.ReadFemPreg()
live = preg[preg["outcome"] == 1]
resp = nsfg.ReadFemResp()
bs = brfss.ReadBrfss()
income = hinc.ReadData()
log_intp_income = hinc2.InterpolateSample(income, log_upper=6.0)
# Q1. Think Stats Chapter 2 Exercise 4 (effect size of Cohen's d)
first_wt = live.loc[preg["birthord"] == 1, "totalwgt_lb"]
other_wt = live.loc[preg["birthord"] != 1, "totalwgt_lb"]
ts.CohenEffectSize(first_wt, other_wt)
# Q2. Think Stats Chapter 3 Exercise 1 (actual vs. biased)
d = np.diff(np.unique(resp["numkdhh"])).min()
left_of_first_bin = resp["numkdhh"].min() - float(d) / 2
right_of_last_bin = resp["numkdhh"].max() + float(d) / 2
plt.clf()
plt.hist(resp["numkdhh"],
bins=np.arange(left_of_first_bin, right_of_last_bin + d, d),
histtype="step",
normed=True,
label="Actual")
plt.hist(resp["numkdhh"],
bins=np.arange(left_of_first_bin, right_of_last_bin + d, d),
histtype="step",
weights=resp["numkdhh"],
def PregnancyLengthDifferences(first, other):
return thinkstats2.CohenEffectSize(first.prglngth, other.prglngth)
new_pmf[val] *= val
new_pmf.Normalize()
return new_pmf
preg = nsfg.ReadFemPreg()
resp = nsfg.ReadFemResp()
live, firsts, others = first.MakeFrames()
#--- Chapter2 Ex4
wgt_live = live.totalwgt_lb.dropna()
wgt_first = firsts.totalwgt_lb.dropna()
wgt_other = others.totalwgt_lb.dropna()
mean_diff = 100 * (wgt_first.mean() - wgt_other.mean()) / wgt_live.mean()
wgt_cohend = thinkstats2.CohenEffectSize(wgt_first, wgt_other)
plen_cohend = thinkstats2.CohenEffectSize(firsts.prglngth, others.prglngth)
print('Difference in relative mean:', mean_diff)
print('Cohen\'s d for total weight in lbs:', wgt_cohend)
print('Cohen\'s d for pregnancy length in weeks:', plen_cohend)
#--- Chapter3 Ex1
actual_pmf = thinkstats2.Pmf(resp.numkdhh, label='actual')
biased_pmf = BiasPmf(actual_pmf, label='biased')
thinkplot.PrePlot(2)
actual_hist = thinkplot.Pmf(actual_pmf)
biased_hist = thinkplot.Pmf(biased_pmf)
thinkplot.Show(xlabel='#kids in household', ylabel='PMF')
print('Actual Mean:', actual_pmf.Mean())
print('Biased Mean:', biased_pmf.Mean())
def preg_length_comparison(live, firsts, others):
# Utilizing existing function from thinkstats2
d = thinkstats2.CohenEffectSize(firsts['prglngth'], others['prglngth'])
return print('Preg Length Cohen Effect Size:', d)
def WeightDifference(firsts, others):
print(firsts.totalwgt_lb.mean() - others.totalwgt_lb.mean())
print(thinkstats2.CohenEffectSize(firsts.totalwgt_lb, others.totalwgt_lb))
def weight_comparison(live, firsts, others):
# Utilizing existing function from thinkstats2
d = thinkstats2.CohenEffectSize(firsts['totalwgt_lb'], others['totalwgt_lb'])
return print('Weight Cohen Effect Size:', d)
def COHEN(Frame1, Frame2):
CD = (Frame1.mean() - Frame2.mean())
SD = ((Frame1.std()**2 + Frame2.std()**2) / 2)**.5
CD = CD / SD
return CD
import thinkstats2, first
live, firsts, others = first.MakeFrames(
) #Gets data from dataframe python. taken from solutions obvs
TotalW = live.totalwgt_lb
FirstW = firsts.totalwgt_lb
OtherW = others.totalwgt_lb
MyCohen = COHEN(FirstW, OtherW)
NotMyCohen = thinkstats2.CohenEffectSize(FirstW, OtherW)
FirstPMean = firsts.prglngth.mean()
OtherPMean = others.prglngth.mean()
print("the mean weight of first babies is " + str(FirstW.mean()) + " lbs")
print("the mean weight of other babies is " + str(OtherW.mean()) + " lbs")
print("the Cohen's d between the two sets is " + str(MyCohen))
print("the mean pregnancy length of first babies was " + str(FirstPMean) +
" weeks")
print("the mean pregnancy length of other babies was " + str(OtherPMean) +
" weeks")
print("the difference between the pregnancy lengths was " +
str(FirstPMean - OtherPMean))
