def Update(self, data):
a_sat, b_sat = data
a_like = thinkbayes2.PmfProbGreater(a_sat, b_sat)
b_like = thinkbayes2.PmfProbLess(a_sat, b_sat)
c_like = thinkbayes2.PmfProbEqual(a_sat, b_sat)
a_like += c_like / 2
b_like += c_like / 2
self.Mult('A', a_like)
self.Mult('B', b_like)
self.Normalize()
def ComparePosteriorPredictive(a_sat, b_sat):
"""Compares the predictive distributions of raw scores.
a_sat: posterior distribution
b_sat:
"""
a_pred = a_sat.MakePredictiveDist()
b_pred = b_sat.MakePredictiveDist()
#thinkplot.Clf()
#thinkplot.Pmfs([a_pred, b_pred])
#thinkplot.Show()
a_like = thinkbayes2.PmfProbGreater(a_pred, b_pred)
b_like = thinkbayes2.PmfProbLess(a_pred, b_pred)
c_like = thinkbayes2.PmfProbEqual(a_pred, b_pred)
print('Posterior predictive')
print('A', a_like)
print('B', b_like)
print('C', c_like)
def main():
#ReadHockeyData()
#return
formats = ['pdf', 'eps']
suite1 = Hockey('bruins')
suite2 = Hockey('canucks')
thinkplot.Clf()
thinkplot.PrePlot(num=2)
thinkplot.Pmf(suite1)
thinkplot.Pmf(suite2)
thinkplot.Save(root='hockey0',
xlabel='Goals per game',
ylabel='Probability',
formats=formats)
suite1.UpdateSet([0, 2, 8, 4])
suite2.UpdateSet([1, 3, 1, 0])
thinkplot.Clf()
thinkplot.PrePlot(num=2)
thinkplot.Pmf(suite1)
thinkplot.Pmf(suite2)
thinkplot.Save(root='hockey1',
xlabel='Goals per game',
ylabel='Probability',
formats=formats)
goal_dist1 = MakeGoalPmf(suite1)
goal_dist2 = MakeGoalPmf(suite2)
thinkplot.Clf()
thinkplot.PrePlot(num=2)
thinkplot.Pmf(goal_dist1)
thinkplot.Pmf(goal_dist2)
thinkplot.Save(root='hockey2',
xlabel='Goals',
ylabel='Probability',
formats=formats)
time_dist1 = MakeGoalTimePmf(suite1)
time_dist2 = MakeGoalTimePmf(suite2)
print('MLE bruins', suite1.MaximumLikelihood())
print('MLE canucks', suite2.MaximumLikelihood())
thinkplot.Clf()
thinkplot.PrePlot(num=2)
thinkplot.Pmf(time_dist1)
thinkplot.Pmf(time_dist2)
thinkplot.Save(root='hockey3',
xlabel='Games until goal',
ylabel='Probability',
formats=formats)
diff = goal_dist1 - goal_dist2
p_win = diff.ProbGreater(0)
p_loss = diff.ProbLess(0)
p_tie = diff.prob(0)
print(p_win, p_loss, p_tie)
p_overtime = thinkbayes2.PmfProbLess(time_dist1, time_dist2)
p_adjust = thinkbayes2.PmfProbEqual(time_dist1, time_dist2)
p_overtime += p_adjust / 2
print('p_overtime', p_overtime)
print(p_overtime * p_tie)
p_win += p_overtime * p_tie
print('p_win', p_win)
# win the next two
p_series = p_win**2
# split the next two, win the third
p_series += 2 * p_win * (1-p_win) * p_win
print('p_series', p_series)