raw_data = generate_data()
df = transform_data(raw_data)
# Table D.4.
# Marked point process (8.14) as mcf1 and cumulative cost process (8.12) as mcf2
lt1 = mcf(df, robust=True, positive=False)
mcf1 = mcfcost(df, mcf_compound=lt1, robust=False, positive=False)
mcf2 = mcfcost(df, robust=True, positive=False)
# Plot data
plot_data(df, alpha=0.5, marker='.', plot_costs=False)
# Plot MCFs
fig, (ax1, ax2) = plt.subplots(1, 2, sharex=True, sharey=False)
plot_mcf(mcf1, ax=ax1, cost=True, label='Events * Cost')
plot_mcf(mcf2, ax=ax2, cost=True, label='Cost')
# Table 8.1.
mcf1['E[C].Std'] = np.sqrt(mcf1['E[C].Var'])
mcf2['E[C].Std'] = np.sqrt(mcf2['E[C].Var'])
mcf1 = mcf1[['Time', 'E[C]', 'E[C].Std']]
mcf2 = mcf2[['Time', 'E[C]', 'E[C].Std']]
mcf1 = mcf1.set_index('Time').reindex([0.50, 1.00, 1.50, 2.00, 2.50],
method='nearest')
mcf2 = mcf2.set_index('Time').reindex([0.50, 1.00, 1.50, 2.00, 2.50],
method='nearest')
mcf1.rename(columns={
'E[C]': 'EST. (8.14)',
'E[C].Std': 'S.E. (8.15)'
},
covariate = 'Drug'
cohort1, cohort2 = 'Placebo', 'Thiotepa'
group_df = df.set_index(covariate)
group_lt = df.groupby(covariate).apply(lambda sfr: mcf(sfr, robust=True, positive=False))
df1, df2 = group_df.ix[cohort1], group_df.ix[cohort2]
lt1, lt2 = group_lt.ix[cohort1], group_lt.ix[cohort2]
lt_diff = mcfdiff(lt1, lt2)
# Plot, no stratification
fig, (ax1, ax2) = plt.subplots(2, 1)
plot_data(df, ax=ax1)
plot_mcf(lt, ax=ax2)
plt.tight_layout()
# Plot, stratified
fig, (ax1, ax2) = plt.subplots(2, 1)
plot_datas([(cohort1, df1), (cohort2, df2)], ax=ax1)
plot_mcfs([(cohort1, lt1), (cohort2, lt2)], ax=ax2)
# Comparison plot
fig, ax1 = plt.subplots(1, 1)
plot_mcfdiff(lt_diff, ax=ax1, label='%s vs. %s' % (cohort1, cohort2))
plt.tight_layout()
# Compute p-values
p_value0 = logrank(df1, df2)
p_value1 = mcfequal(df1, df2)
sep=';') #, usecols=[0,1,2])
cohort1, cohort2 = 'Male', 'Female'
df1 = transform_population(
raw_data[raw_data['Sex'] == cohort1]).fillna(cohort1)
df2 = transform_population(
raw_data[raw_data['Sex'] == cohort2]).fillna(cohort2)
df = pd.concat((df1, df2))
lt1 = mcf(df1, robust=False, positive=False)
lt2 = mcf(df2, robust=False, positive=False)
lt_diff = mcfdiff(lt1, lt2)
# Plot, stratified
fig, (ax1, ax2, ax3) = plt.subplots(1, 3, sharey=False, sharex=True)
plot_mcf(lt1, ax=ax1, interval=True, CI=True, label=cohort1)
plot_mcfs([(cohort1, lt1), (cohort2, lt2)],
ax=ax2,
interval=True,
CI=False)
plot_mcfdiff(lt_diff, ax=ax3, label='%s vs. %s' % (cohort1, cohort2))
ax1.set_ylim([0, 2])
ax2.set_ylim([0, 2])
ax3.set_ylim([-1, 1])
plt.xlim([0, 60])
# Compute p-values
p_value1 = mcfequal(df1, df2)
print "p-value: %.3f" % (p_value1)
plt.show()
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from data_operations import transform_population, mcf, mcfdiff, mcfequal,\
plot_data, plot_datas, plot_mcf, plot_mcfs, plot_mcfdiff, plot_mcfdiffs, population_reverse
if __name__ == '__main__':
# Defrost Control Example
raw_data = pd.read_csv('datas/defrost_interval.csv', sep=';')#, usecols=[0,1,2])
df = transform_population(raw_data)
lt1 = mcf(df, robust=False, positive=False, interval=True)
print df.head()
fr = population_reverse(df)
plot_data(fr, alpha=0.005)
fig, (ax1) = plt.subplots(1, 1, sharex=True, sharey=True)
plot_mcf(lt1, ax=ax1, interval=True)
plt.ylim([0.0, 0.1])
plt.show()
#lt_cost = mcf_cost(fr, mcf_compound=lt)
#print lt_cost.head()
#plot_cost(mcf_cost)
df_sessions = transform_data(raw_data[(raw_data['Type'] == 0) |
(raw_data['Event'] == 0)])
df_purchases = transform_data(raw_data[(raw_data['Type'] == 1) |
(raw_data['Event'] == 0)])
mcf_sessions = mcfcost(df_sessions, robust=True, positive=True)
mcf_purchases = mcfcost(df_purchases, robust=True, positive=True)
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2,
2,
sharex=True,
sharey=False)
plot_data(df_sessions, ax=ax1, marker='.', label='Session', alpha=0.2)
plot_data(df_purchases, ax=ax2, marker='*', label='Purchase', alpha=0.2)
plot_mcf(mcf_sessions, ax=ax3, cost=True).set_ylabel('Playtime')
plot_mcf(mcf_purchases, ax=ax4, cost=True).set_ylabel('Lifetime Value')
T = np.linspace(0, max_date, 100)
E_playtime = playtime_dist.mean() * play_rate / churn_rate * (
1 - np.exp(-churn_rate * T))
E_purchases = purchase_dist.mean() * purchase_rate / churn_rate * (
1 - np.exp(-churn_rate * T))
ax3.plot(T,
E_playtime,
linestyle='-',
alpha=0.5,
color='black',
label='True')
ax4.plot(T,
E_purchases,
import matplotlib.pyplot as plt
from data_operations import transform_population, mcf, mcfdiff, mcfequal,\
plot_data, plot_datas, plot_mcf, plot_mcfs, plot_mcfdiff, plot_mcfdiffs, population_reverse
if __name__ == '__main__':
# Compressor Example
raw_data = pd.read_csv('datas/compressor_population.csv',
sep=';') #, usecols=[0,1,2])
#raw_data = raw_data.groupby('Time', as_index=False).sum()
df = transform_population(raw_data)
print population_reverse(df)
group_df = raw_data.groupby('Building').apply(
lambda sdata: population_reverse(transform_population(sdata)))
dfs = list(group_df.groupby(level=0))
lt1 = mcf(df, robust=False, positive=False)
lt2 = mcf(df, robust=False, positive=True)
plot_datas(dfs, alpha=0.1)
fig, (ax1, ax2) = plt.subplots(1, 2, sharex=True, sharey=True)
plot_mcf(lt1, ax=ax1, label='Normal limits')
plot_mcf(lt2, ax=ax2, label='Normal limits (positive)')
plt.ylim([-0.005, 0.1])
plt.show()
#lt_cost = mcf_cost(fr, mcf_compound=lt)
#print lt_cost.head()
#plot_cost(mcf_cost)
# Fan Example
raw_data = pd.read_csv('datas/fan_population.csv',
sep=';') #, usecols=[0,1,2])
raw_data.rename(columns={'Cost': 'Costs'}, inplace=True)
df = transform_population(raw_data)
fr = population_reverse(df)
fr2 = population_reverse_costs(df)
lt1 = mcf(df, robust=False, positive=False)
lt2 = mcf(df, robust=False, positive=True)
mcf1 = mcfcost(df, mcf_compound=lt1, robust=False, positive=False)
mcf2 = mcfcost(df, mcf_compound=lt2, robust=False, positive=True)
plot_data(fr2, alpha=0.5)
fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2,
2,
sharex=True,
sharey=False)
plot_mcf(lt1, ax=ax1, label='Normal limits')
plot_mcf(lt2, ax=ax2, label='Normal limits (positive)')
plot_mcf(mcf1, ax=ax3, cost=True, label='Normal limits')
plot_mcf(mcf2, ax=ax4, cost=True, label='Normal limits (positive)')
plt.show()
ax1.set_ylim([-0.01, 0.51])
ax2.set_ylim([-0.01, 0.51])
#lt_cost = mcf_cost(fr, mcf_compound=lt)
#print lt_cost.head()
#plot_cost(mcf_cost)