def causal_check():
# Check IPTW plots
data = load_sample_data(False)
data[['cd4_rs1', 'cd4_rs2']] = spline(data,
'cd40',
n_knots=3,
term=2,
restricted=True)
data[['age_rs1', 'age_rs2']] = spline(data,
'age0',
n_knots=3,
term=2,
restricted=True)
ipt = IPTW(data, treatment='art', stabilized=True)
ipt.regression_models(
'male + age0 + age_rs1 + age_rs2 + cd40 + cd4_rs1 + cd4_rs2 + dvl0')
ipt.fit()
ipt.plot_love()
plt.tight_layout()
plt.show()
ipt.plot_kde()
plt.show()
ipt.plot_kde(measure='logit')
plt.show()
ipt.plot_boxplot()
plt.show()
ipt.plot_boxplot(measure='logit')
plt.show()
# Check SurvivalGFormula plots
df = load_sample_data(False).drop(columns=['cd4_wk45'])
df['t'] = np.round(df['t']).astype(int)
df = pd.DataFrame(np.repeat(df.values, df['t'], axis=0),
columns=df.columns)
df['t'] = df.groupby('id')['t'].cumcount() + 1
df.loc[((df['dead'] == 1) & (df['id'] != df['id'].shift(-1))), 'd'] = 1
df['d'] = df['d'].fillna(0)
df['t_sq'] = df['t']**2
df['t_cu'] = df['t']**3
sgf = SurvivalGFormula(df,
idvar='id',
exposure='art',
outcome='d',
time='t')
sgf.outcome_model(
model='art + male + age0 + cd40 + dvl0 + t + t_sq + t_cu')
sgf.fit(treatment='all')
sgf.plot()
plt.show()
sgf.plot(c='r', linewidth=3, alpha=0.8)
plt.show()
def test_unstabilized_weights(self):
df = load_sample_data(False)
ipm = IPMW(df, missing_variable='dead', stabilized=False)
ipm.regression_models(model_denominator='male + age0 + dvl0 + cd40')
ipm.fit()
npt.assert_allclose(np.mean(ipm.Weight), 1.0579602715)
npt.assert_allclose(np.std(ipm.Weight, ddof=1), 0.021019729152)
def test_weighted_model(self):
df = load_sample_data(False).drop(columns=['dead'])
df['age_sq'] = df['age0'] ** 2
df['age_cu'] = df['age0'] ** 3
df['cd4_sq'] = df['cd40'] ** 2
df['cd4_cu'] = df['cd40'] ** 3
ipmw = IPMW(df, missing_variable='cd4_wk45')
ipmw.regression_models('art + male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0',
print_results=False)
ipmw.fit()
df['ipcw'] = ipmw.Weight
snm = GEstimationSNM(df.dropna(), exposure='art', outcome='cd4_wk45', weights='ipcw')
snm.exposure_model('male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0', print_results=False)
snm.structural_nested_model('art')
snm.fit()
npt.assert_allclose(snm.psi, [244.379181], atol=1e-5)
snm = GEstimationSNM(df.dropna(), exposure='art', outcome='cd4_wk45', weights='ipcw')
snm.exposure_model('male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0', print_results=False)
snm.structural_nested_model('art')
snm.fit(solver='search')
npt.assert_allclose(snm.psi, [244.379181], atol=1e-5)
def test_missing_w_weights(self):
df = load_sample_data(False).drop(columns=['dead'])
df['age_sq'] = df['age0'] ** 2
df['age_cu'] = df['age0'] ** 3
df['cd4_sq'] = df['cd40'] ** 2
df['cd4_cu'] = df['cd40'] ** 3
df['weight'] = 2
snm = GEstimationSNM(df, exposure='art', outcome='cd4_wk45', weights='weight')
snm.exposure_model('male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0', print_results=False)
snm.missing_model('art + male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0',
stabilized=False, print_results=False)
snm.structural_nested_model('art')
snm.fit()
npt.assert_allclose(snm.psi, [244.379181], atol=1e-5)
snm = GEstimationSNM(df, exposure='art', outcome='cd4_wk45')
snm.exposure_model('male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0', print_results=False)
snm.missing_model('art + male + age0 + age_sq + age_cu + cd40 + cd4_sq + cd4_cu + dvl0',
stabilized=False, print_results=False)
snm.structural_nested_model('art')
snm.fit(solver='search')
npt.assert_allclose(snm.psi, [244.379181], atol=1e-5)
def test_error_numerator_with_unstabilized(self):
df = load_sample_data(False)
ipm = IPMW(df, missing_variable='dead', stabilized=False)
with pytest.raises(ValueError):
ipm.regression_models(
model_denominator='male + age0 + dvl0 + cd40',
model_numerator='male')
def test_stabilized_weights(self):
df = load_sample_data(False)
ipm = IPMW(df, missing_variable='dead', stabilized=True)
ipm.regression_models(model_denominator='male + age0 + dvl0 + cd40')
ipm.fit()
npt.assert_allclose(np.mean(ipm.Weight), 0.99993685627)
npt.assert_allclose(np.std(ipm.Weight, ddof=1), 0.019866910369)
def data_b(self):
df = load_sample_data(False).drop(columns=['cd4_wk45']).dropna()
df['age_sq'] = df['age0']**2
df['age_cu'] = df['age0']**3
df['cd4_sq'] = df['cd40']**2
df['cd4_cu'] = df['cd40']**3
return df
def test_match_openepi_sampledata(self):
oe_irr = -0.001055
oe_ci = -0.003275, 0.001166
df = ze.load_sample_data(False)
ird = IncidenceRateDifference()
ird.fit(df, exposure='art', outcome='dead', time='t')
npt.assert_allclose(ird.incidence_rate_difference[1], oe_irr, atol=1e-5)
rf = ird.results
npt.assert_allclose(rf.loc[rf.index == '1'][['IRD_LCL', 'IRD_UCL']], [oe_ci], atol=1e-5)
def test_mc_detect_censoring(self):
df = load_sample_data(timevary=True)
not_censored = np.where((df['id'] != df['id'].shift(-1)) & (df['dead'] == 0), 0, 1)
not_censored = np.where(df['out'] == np.max(df['out']), 1, not_censored)
g = MonteCarloGFormula(df, idvar='id', exposure='art', outcome='dead', time_in='enter', time_out='out')
npt.assert_equal(np.array(g.gf['__uncensored__']), not_censored)
def data(self):
df = load_sample_data(False).drop(columns=['cd4_wk45'])
df['t'] = np.round(df['t']).astype(int)
df = pd.DataFrame(np.repeat(df.values, df['t'], axis=0), columns=df.columns)
df['t'] = df.groupby('id')['t'].cumcount() + 1
df.loc[((df['dead'] == 1) & (df['id'] != df['id'].shift(-1))), 'd'] = 1
df['d'] = df['d'].fillna(0)
df['t_sq'] = df['t'] ** 2
df['t_cu'] = df['t'] ** 3
return df.drop(columns=['dead'])
def test_match_sas_sampledata(self):
sas_rr = -0.045129870
sas_se = 0.042375793
sas_ci = -0.128184899, 0.037925158
df = ze.load_sample_data(False)
rd = RiskDifference()
rd.fit(df, exposure='art', outcome='dead')
npt.assert_allclose(rd.risk_difference[1], sas_rr)
rf = rd.results
npt.assert_allclose(rf.loc[rf.index == '1'][['RD_LCL', 'RD_UCL']], [sas_ci])
npt.assert_allclose(rf.loc[rf.index == '1'][['SD(RD)']], sas_se)
def test_match_sas_sampledata(self):
sas_rd = 0.742118331
sas_se = 0.312612740
sas_ci = 0.402139480, 1.369523870
df = ze.load_sample_data(False)
rr = RiskRatio()
rr.fit(df, exposure='art', outcome='dead')
npt.assert_allclose(rr.risk_ratio[1], sas_rd, rtol=1e-5)
rf = rr.results
npt.assert_allclose(rf.loc[rf.index == '1'][['RR_LCL', 'RR_UCL']], [sas_ci], rtol=1e-5)
npt.assert_allclose(rf.loc[rf.index == '1'][['SD(RR)']], sas_se, rtol=1e-5)
def test_match_sas_sampledata(self):
sas_or = 0.7036
sas_se = 0.361479191
sas_ci = 0.3465, 1.4290
df = ze.load_sample_data(False)
ord = OddsRatio()
ord.fit(df, exposure='art', outcome='dead')
npt.assert_allclose(ord.odds_ratio[1], sas_or, rtol=1e-4)
rf = ord.results
npt.assert_allclose(rf.loc[rf.index == '1'][['OR_LCL', 'OR_UCL']], [sas_ci], rtol=1e-3)
npt.assert_allclose(rf.loc[rf.index == '1'][['SD(OR)']], sas_se, rtol=1e-4)
def test_match_sas_sampledata(self):
sas_irr = 0.753956
sas_se = 0.336135409
sas_ci = 0.390146, 1.457017
df = ze.load_sample_data(False)
irr = IncidenceRateRatio()
irr.fit(df, exposure='art', outcome='dead', time='t')
npt.assert_allclose(irr.incidence_rate_ratio[1], sas_irr, rtol=1e-5)
rf = irr.results
npt.assert_allclose(rf.loc[rf.index == '1'][['IRR_LCL', 'IRR_UCL']], [sas_ci], rtol=1e-5)
npt.assert_allclose(rf.loc[rf.index == '1'][['SD(IRR)']], sas_se, rtol=1e-5)
def sdata():
df = load_sample_data(False)
df[['cd4_rs1', 'cd4_rs2']] = spline(df,
'cd40',
n_knots=3,
term=2,
restricted=True)
df[['age_rs1', 'age_rs2']] = spline(df,
'age0',
n_knots=3,
term=2,
restricted=True)
return df.drop(columns=['cd4_wk45'])
def mcf(self):
df = ze.load_sample_data(False)
df[['cd4_rs1', 'cd4_rs2']] = ze.spline(df,
'cd40',
n_knots=3,
term=2,
restricted=True)
df[['age_rs1', 'age_rs2']] = ze.spline(df,
'age0',
n_knots=3,
term=2,
restricted=True)
return df.drop(columns=['dead'])
def test_match_inverse_of_risk_difference(self):
df = ze.load_sample_data(False)
rd = RiskDifference()
rd.fit(df, exposure='art', outcome='dead')
nnt = NNT()
nnt.fit(df, exposure='art', outcome='dead')
npt.assert_allclose(nnt.number_needed_to_treat[1], 1/rd.risk_difference[1])
rf = rd.results
nf = nnt.results
npt.assert_allclose(nf.loc[nf.index == '1'][['NNT_LCL', 'NNT_UCL']],
1 / rf.loc[rf.index == '1'][['RD_LCL', 'RD_UCL']])
npt.assert_allclose(nf.loc[nf.index == '1'][['SD(RD)']], rf.loc[rf.index == '1'][['SD(RD)']])
def mc_gformula_check():
df = load_sample_data(timevary=True)
df['lag_art'] = df['art'].shift(1)
df['lag_art'] = np.where(df.groupby('id').cumcount() == 0, 0, df['lag_art'])
df['lag_cd4'] = df['cd4'].shift(1)
df['lag_cd4'] = np.where(df.groupby('id').cumcount() == 0, df['cd40'], df['lag_cd4'])
df['lag_dvl'] = df['dvl'].shift(1)
df['lag_dvl'] = np.where(df.groupby('id').cumcount() == 0, df['dvl0'], df['lag_dvl'])
df[['age_rs0', 'age_rs1', 'age_rs2']] = spline(df, 'age0', n_knots=4, term=2, restricted=True) # age spline
df['cd40_sq'] = df['cd40'] ** 2 # cd4 baseline cubic
df['cd40_cu'] = df['cd40'] ** 3
df['cd4_sq'] = df['cd4'] ** 2 # cd4 current cubic
df['cd4_cu'] = df['cd4'] ** 3
df['enter_sq'] = df['enter'] ** 2 # entry time cubic
df['enter_cu'] = df['enter'] ** 3
g = TimeVaryGFormula(df, idvar='id', exposure='art', outcome='dead', time_in='enter', time_out='out')
exp_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + cd4 + cd4_sq +
cd4_cu + dvl + enter + enter_sq + enter_cu'''
g.exposure_model(exp_m, restriction="g['lag_art']==0")
out_m = '''art + male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + cd4 +
cd4_sq + cd4_cu + dvl + enter + enter_sq + enter_cu'''
g.outcome_model(out_m, restriction="g['drop']==0")
dvl_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + lag_cd4 +
lag_dvl + lag_art + enter + enter_sq + enter_cu'''
g.add_covariate_model(label=1, covariate='dvl', model=dvl_m, var_type='binary')
cd4_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + lag_cd4 +
lag_dvl + lag_art + enter + enter_sq + enter_cu'''
cd4_recode_scheme = ("g['cd4'] = np.maximum(g['cd4'],1);"
"g['cd4_sq'] = g['cd4']**2;"
"g['cd4_cu'] = g['cd4']**3")
g.add_covariate_model(label=2, covariate='cd4', model=cd4_m,recode=cd4_recode_scheme, var_type='continuous')
g.fit(treatment="((g['art']==1) | (g['lag_art']==1))",
lags={'art': 'lag_art',
'cd4': 'lag_cd4',
'dvl': 'lag_dvl'},
sample=10000, t_max=None,
in_recode=("g['enter_sq'] = g['enter']**2;"
"g['enter_cu'] = g['enter']**3"))
gf = g.predicted_outcomes
gfs = gf.loc[gf.uid_g_zepid != gf.uid_g_zepid.shift(-1)].copy()
kmn = KaplanMeierFitter()
kmn.fit(durations=gfs['out'], event_observed=gfs['dead'])
kmo = KaplanMeierFitter()
kmo.fit(durations=df['out'], event_observed=df['dead'], entry=df['enter'])
plt.step(kmn.event_table.index, 1 - kmn.survival_function_, c='g', where='post', label='Natural')
plt.step(kmo.event_table.index, 1 - kmo.survival_function_, c='k', where='post', label='True')
plt.legend()
plt.show()
def causal_check():
# 9) Check IPTW plots
data = load_sample_data(False)
data[['cd4_rs1', 'cd4_rs2']] = spline(data, 'cd40', n_knots=3, term=2, restricted=True)
data[['age_rs1', 'age_rs2']] = spline(data, 'age0', n_knots=3, term=2, restricted=True)
ipt = IPTW(data, treatment='art', stabilized=True)
ipt.regression_models('male + age0 + age_rs1 + age_rs2 + cd40 + cd4_rs1 + cd4_rs2 + dvl0')
ipt.fit()
ipt.plot_love()
plt.tight_layout()
plt.show()
ipt.plot_kde()
plt.show()
ipt.plot_kde(measure='logit')
plt.show()
ipt.plot_boxplot()
plt.show()
ipt.plot_boxplot(measure='logit')
plt.show()
def test_complete_mc_procedure_completes(self):
df = load_sample_data(timevary=True)
df['lag_art'] = df['art'].shift(1)
df['lag_art'] = np.where(df.groupby('id').cumcount() == 0, 0, df['lag_art'])
df['lag_cd4'] = df['cd4'].shift(1)
df['lag_cd4'] = np.where(df.groupby('id').cumcount() == 0, df['cd40'], df['lag_cd4'])
df['lag_dvl'] = df['dvl'].shift(1)
df['lag_dvl'] = np.where(df.groupby('id').cumcount() == 0, df['dvl0'], df['lag_dvl'])
df[['age_rs0', 'age_rs1', 'age_rs2']] = spline(df, 'age0', n_knots=4, term=2, restricted=True) # age spline
df['cd40_sq'] = df['cd40'] ** 2
df['cd40_cu'] = df['cd40'] ** 3
df['cd4_sq'] = df['cd4'] ** 2
df['cd4_cu'] = df['cd4'] ** 3
df['enter_sq'] = df['enter'] ** 2
df['enter_cu'] = df['enter'] ** 3
g = MonteCarloGFormula(df, idvar='id', exposure='art', outcome='dead', time_in='enter', time_out='out')
exp_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + cd4 + cd4_sq +
cd4_cu + dvl + enter + enter_sq + enter_cu'''
g.exposure_model(exp_m, restriction="g['lag_art']==0")
out_m = '''art + male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + cd4 +
cd4_sq + cd4_cu + dvl + enter + enter_sq + enter_cu'''
g.outcome_model(out_m, restriction="g['drop']==0")
dvl_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + lag_cd4 +
lag_dvl + lag_art + enter + enter_sq + enter_cu'''
g.add_covariate_model(label=1, covariate='dvl', model=dvl_m, var_type='binary')
cd4_m = '''male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + lag_cd4 +
lag_dvl + lag_art + enter + enter_sq + enter_cu'''
cd4_recode_scheme = ("g['cd4'] = np.maximum(g['cd4'],1);"
"g['cd4_sq'] = g['cd4']**2;"
"g['cd4_cu'] = g['cd4']**3")
g.add_covariate_model(label=2, covariate='cd4', model=cd4_m, recode=cd4_recode_scheme, var_type='continuous')
cens_m = """male + age0 + age_rs0 + age_rs1 + age_rs2 + cd40 + cd40_sq + cd40_cu + dvl0 + lag_cd4 +
lag_dvl + lag_art + enter + enter_sq + enter_cu"""
g.censoring_model(cens_m)
g.fit(treatment="((g['art']==1) | (g['lag_art']==1))",
lags={'art': 'lag_art',
'cd4': 'lag_cd4',
'dvl': 'lag_dvl'},
sample=5000, t_max=None,
in_recode=("g['enter_sq'] = g['enter']**2;"
"g['enter_cu'] = g['enter']**3"))
assert isinstance(g.predicted_outcomes, type(pd.DataFrame()))
def measures_check():
# 7) Check measures plots
data_set = load_sample_data(False)
rr = RiskRatio()
rr.fit(data_set, exposure='art', outcome='dead')
rr.plot(fmt='*', ecolor='r', barsabove=True, markersize=25)
plt.show()
rd = RiskDifference()
rd.fit(data_set, exposure='art', outcome='dead')
rd.plot()
plt.show()
ord = OddsRatio()
ord.fit(data_set, exposure='art', outcome='dead')
ord.plot()
plt.show()
irr = IncidenceRateRatio()
irr.fit(data_set, exposure='art', outcome='dead', time='t')
irr.plot()
plt.show()
ird = IncidenceRateDifference()
ird.fit(data_set, exposure='art', outcome='dead', time='t')
ird.plot()
plt.show()
def test_match_sas_weights(self):
sas_w_mean = 0.9993069
sas_w_max = 1.7980410
sas_w_min = 0.8986452
df = load_sample_data(timevary=True)
df['cd40_q'] = df['cd40'] ** 2
df['cd40_c'] = df['cd40'] ** 3
df['cd4_q'] = df['cd4'] ** 2
df['cd4_c'] = df['cd4'] ** 3
df['enter_q'] = df['enter'] ** 2
df['enter_c'] = df['enter'] ** 3
df['age0_q'] = df['age0'] ** 2
df['age0_c'] = df['age0'] ** 3
ipc = IPCW(df, idvar='id', time='enter', event='dead')
cmodeln = 'enter + enter_q + enter_c'
cmodeld = '''enter + enter_q + enter_c + male + age0 + age0_q + age0_c + dvl0 + cd40 +
cd40_q + cd40_c + dvl + cd4 + cd4_q + cd4_c'''
ipc.regression_models(model_denominator=cmodeld, model_numerator=cmodeln)
ipc.fit()
cw = ipc.Weight
npt.assert_allclose(np.mean(cw), sas_w_mean)
npt.assert_allclose(np.max(cw), sas_w_max)
npt.assert_allclose(np.min(cw), sas_w_min)
def test_return_pandas(self):
df = ze.load_sample_data(False)
assert isinstance(df, type(pd.DataFrame()))
def test_correct_nobs_timevary(self):
df = ze.load_sample_data(True)
assert df.shape[0] == 27382
def test_correct_ncols_timevary(self):
df = ze.load_sample_data(True)
assert df.shape[1] == 12
def test_missing_in_timefixed(self):
df = ze.load_sample_data(False)
assert df.shape[0] != df.dropna().shape[0]
def test_correct_ncols_timefixed(self):
df = ze.load_sample_data(False)
assert df.shape[1] == 9
def test_error_too_many_model(self):
df = load_sample_data(False)
ipm = IPMW(df, missing_variable='dead')
with pytest.raises(ValueError):
ipm.regression_models(
model_denominator=['male + age0', 'male + age0 + dvl0'])
def graphics_check():
# L'Abbe Plots
labbe_plot()
plt.show()
labbe_plot(r1=[0.25, 0.5], r0=[0.1, 0.2], color='r')
plt.show()
labbe_plot(r1=[0.3, 0.5], r0=[0.2, 0.7], scale='additive', marker='+', linestyle='')
plt.show()
labbe_plot(r1=[0.3, 0.5], r0=[0.2, 0.7], scale='multiplicative', markersize=10)
plt.show()
# 1) Check EffectMeasurePlot
labs = ['Overall', 'Adjusted', '', '2012-2013', 'Adjusted', '', '2013-2014', 'Adjusted', '', '2014-2015',
'Adjusted']
measure = [np.nan, 0.94, np.nan, np.nan, 1.22, np.nan, np.nan, 0.59, np.nan, np.nan, 1.09]
lower = [np.nan, 0.77, np.nan, np.nan, '0.80', np.nan, np.nan, '0.40', np.nan, np.nan, 0.83]
upper = [np.nan, 1.15, np.nan, np.nan, 1.84, np.nan, np.nan, 0.85, np.nan, np.nan, 1.44]
p = EffectMeasurePlot(label=labs, effect_measure=measure, lcl=lower, ucl=upper)
p.plot(figsize=[7, 4])
plt.show()
# 2) Check Functional form plots
data = load_sample_data(False)
data['age_sq'] = data['age0']**2
functional_form_plot(data, 'dead', var='age0', loess=False)
plt.show()
functional_form_plot(data, 'dead', var='age0', discrete=True, loess=False)
plt.show()
functional_form_plot(data, 'cd40', var='age0', outcome_type='continuous', loess=False)
plt.show()
functional_form_plot(data, 'dead', var='age0', points=True, loess=False)
plt.show()
functional_form_plot(data, 'dead', var='age0', loess=True, loess_value=0.25, discrete=True)
plt.show()
functional_form_plot(data, 'dead', var='age0', f_form='age0 + age_sq', loess=False)
plt.show()
# 3) Check P-value plots
p = pvalue_plot(point=0.23, sd=0.1)
plt.show()
pvalue_plot(point=0.23, sd=0.1, fill=False)
plt.show()
pvalue_plot(point=0.23, sd=0.1, color='r')
plt.show()
pvalue_plot(point=0.23, sd=0.1, null=0.05)
plt.show()
pvalue_plot(point=0.23, sd=0.1, alpha=0.05)
plt.show()
# 4) Check Spaghetti plot
df = load_sample_data(timevary=True)
spaghetti_plot(df, idvar='id', variable='cd4', time='enter')
plt.show()
# 5) Check ROC
df = pd.DataFrame()
df['d'] = [0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1]
df['p'] = [0.1, 0.15, 0.1, 0.7, 0.5, 0.9, 0.95, 0.5, 0.4, 0.8, 0.99, 0.99, 0.89, 0.95]
roc(df, true='d', threshold='p', youden_index=False)
plt.show()
roc(df, true='d', threshold='p', youden_index=True)
plt.show()
# 6) Check Dynamic Risk Plots
a = pd.DataFrame([[0, 0], [1, 0.15], [2, 0.25], [4, 0.345]], columns=['timeline', 'riske']).set_index(
'timeline')
b = pd.DataFrame([[0, 0], [1, 0.2], [1.5, 0.31], [3, 0.345]], columns=['timeline', 'riske']).set_index(
'timeline')
dynamic_risk_plot(a, b, loess=False)
plt.show()
dynamic_risk_plot(a, b, measure='RR', loess=False)
plt.show()
dynamic_risk_plot(a, b, measure='RR', scale='log-transform', loess=False)
plt.show()
dynamic_risk_plot(a, b, measure='RR', scale='log', loess=False)
plt.show()
dynamic_risk_plot(a, b, loess=True, loess_value=0.4)
plt.show()
dynamic_risk_plot(a, b, loess=False, point_color='green', line_color='green')
plt.show()
def test_missing_count2(self):
df = load_sample_data(False)
ipm = IPMW(df, missing_variable='dead', stabilized=True)
ipm.regression_models(model_denominator='art')
assert 517 == np.sum(ipm.df['_observed_indicator_'])
assert 30 == np.sum(1 - ipm.df['_observed_indicator_'])
