def run_diagnostics(self, decimal=3):
"""Run all currently implemented diagnostics for the exposure and outcome models. Each
`run_diagnostics` provides results for all implemented diagnostics for ease of the user. For publication
quality presentations, I recommend calling each diagnostic function individually and utilizing the optional
parameters
Note
----
The plot presented cannot be edited. To edit the plots, call `plot_kde` or `plot_love` directly. Those
functions return an axes object
Parameters
----------
decimal : int, optional
Number of decimal places to display. Default is 3
Returns
-------
None
"""
if not self._fit_outcome_ or not self._fit_exposure_:
raise ValueError("The exposure_model and outcome_model function must be ran before any diagnostics")
# Weight diagnostics
print('\tExposure Model Diagnostics')
self.positivity(decimal=decimal)
print('\n======================================================================')
print(' Standardized Mean Differences')
print('======================================================================')
print(self.standardized_mean_differences().set_index(keys='labels'))
print('======================================================================\n')
# Outcome accuracy diagnostics
print('\tOutcome Model Diagnostics')
v = self._predicted_y_ - self.df[self.outcome]
outcome_accuracy(true=self.df[self.outcome], predicted=self._predicted_y_, decimal=decimal)
df = self.df.copy()
df['_ipw_'] = np.where(df[self.exposure] == 1, 1 / df['_g1_'], 1 / (1 - df['_g1_']))
plt.figure(figsize=[8, 6])
plt.subplot(221)
plot_love(df=df, treatment=self.exposure, weight='_ipw_', formula=self.__mweight)
plt.title("Love Plot")
plt.subplot(223)
plot_kde(df=df, treatment=self.exposure, probability='_g1_')
plt.title("Kernel Density of Propensity Scores")
plt.subplot(222)
plot_kde_accuracy(values=v.dropna(), color='green')
plt.title("Kernel Density of Accuracy")
plt.tight_layout()
plt.show()
def plot_love(self, color_unweighted='r', color_weighted='b', shape_unweighted='o', shape_weighted='o'):
"""Generates a Love-plot to detail covariate balance based on the IPTW weights. Further details on the usage of
this plot are available in Austin PC & Stuart EA 2015 https://onlinelibrary.wiley.com/doi/full/10.1002/sim.6607
The Love plot generates a dashed line at standardized mean difference of 0.10. Ideally, weighted SMD are below
this level. Below 0.20 may also be sufficient. Variables above this level may be unbalanced despite the
weighting procedure. Different functional forms (or approaches like machine learning) may be worth considering
Parameters
----------
color_unweighted : str, optional
Color for the unweighted standardized mean differences. Default is red
color_weighted : str, optional
Color for the weighted standardized mean differences. Default is blue
shape_unweighted : str, optional
Shape of points for the unweighted standardized mean differences. Default is circles
shape_weighted:
Shape of points for the weighted standardized mean differences. Default is circles
Returns
-------
axes
Matplotlib axes of the Love plot
"""
df = self.df.copy()
df['_ipw_'] = np.where(df[self.exposure] == 1, 1 / df['_g1_'], 1 / (1 - df['_g1_']))
ax = plot_love(df=df, treatment=self.exposure, weight='_ipw_', formula=self.__mweight,
color_unweighted=color_unweighted, color_weighted=color_weighted,
shape_unweighted=shape_unweighted, shape_weighted=shape_weighted)
return ax
def plot_love(self,
color_unweighted='r',
color_weighted='b',
shape_unweighted='o',
shape_weighted='o',
iptw_only=True):
"""Generates a Love-plot to detail covariate balance based on the IPTW weights. Further details on the usage of
this plot are available in Austin PC & Stuart EA 2015 https://onlinelibrary.wiley.com/doi/full/10.1002/sim.6607
The Love plot generates a dashed line at standardized mean difference of 0.10. Ideally, weighted SMD are below
this level. Below 0.20 may also be sufficient. Variables above this level may be unbalanced despite the
weighting procedure. Different functional forms (or approaches like machine learning) may be worth considering
Parameters
----------
color_unweighted : str, optional
Color for the unweighted standardized mean differences. Default is red
color_weighted : str, optional
Color for the weighted standardized mean differences. Default is blue
shape_unweighted : str, optional
Shape of points for the unweighted standardized mean differences. Default is circles
shape_weighted:
Shape of points for the weighted standardized mean differences. Default is circles
iptw_only : bool, optional
Whether the diagnostic should be run on IPTW only or the weights multiplied together. Default is IPTW only
Returns
-------
matplotlib axes
"""
if iptw_only:
df = self.df
df['_ipfw_'] = self.iptw
else:
df = self.df
df['_ipfw_'] = self.iptw * self.ipmw
ax = plot_love(df=self.df,
treatment=self.treatment,
weight='_ipfw_',
formula=self.__mdenom,
color_unweighted=color_unweighted,
color_weighted=color_weighted,
shape_unweighted=shape_unweighted,
shape_weighted=shape_weighted)
return ax