causallib.survival.WeightedSurvival#

class WeightedSurvival(weight_model=None, survival_model=None)[source]#

Weighted survival estimator

Weighted survival estimator. :param weight_model: causallib compatible weight model (e.g., IPW) :param survival_model: Three alternatives:

  1. None - compute non-parametric KaplanMeier survival curve

  2. Scikit-Learn estimator (needs to implement predict_proba) - compute parametric curve by fitting a

    time-varying hazards model

  3. lifelines UnivariateFitter - use lifelines fitter to compute survival curves from events and durations

__init__(weight_model=None, survival_model=None)[source]#

Weighted survival estimator. :param weight_model: causallib compatible weight model (e.g., IPW) :param survival_model: Three alternatives:

  1. None - compute non-parametric KaplanMeier survival curve

  2. Scikit-Learn estimator (needs to implement predict_proba) - compute parametric curve by fitting a

    time-varying hazards model

  3. lifelines UnivariateFitter - use lifelines fitter to compute survival curves from events and durations

fit(X, a, t=None, y=None, fit_kwargs=None)[source]#

Fits internal weight module (e.g. IPW module, adversarial weighting, etc).

Parameters:

  • X (pandas.DataFrame) – Baseline covariate matrix of size (num_subjects, num_features).

  • a (pandas.Series) – Treatment assignment of size (num_subjects,).

  • t (pandas.Series) – NOT USED (for compatibility only)

  • y (pandas.Series) – NOT USED (for compatibility only)

  • fit_kwargs (dict) – Optional kwargs for fit call of survival model (NOT USED, since fit call of survival model occurs in ‘estimate_population_outcome’ rather than here)

Returns:

self

estimate_population_outcome(X, a, t, y, timeline_start=None, timeline_end=None)[source]#

Returns population averaged survival curves.

Parameters:

  • X (pandas.DataFrame) – Baseline covariate matrix of size (num_subjects, num_features).

  • a (pandas.Series) – Treatment assignment of size (num_subjects,).

  • t (pandas.Series|int) – Followup durations, size (num_subjects,).

  • y (pandas.Series) – Observed outcome (1) or right censoring event (0), size (num_subjects,).

  • timeline_start (int) – Common start time-step. If provided, will generate survival curves starting from ‘timeline_start’ for all patients. If None, will predict from first observed event.

  • timeline_end (int) – Common end time-step. If provided, will generate survival curves up to ‘timeline_end’ for all patients. If None, will predict up to last observed event.

Returns:

with timestep index, treatment values as columns and survival as entries

Return type:

pandas.DataFrame