xgbse.extrapolation

extrapolate_constant_risk(survival, final_time, intervals, lags=-1)

Extrapolate a survival curve assuming constant risk.

Parameters:

Name	Type	Description	Default
survival	pd.DataFrame	A dataframe of survival probabilities for all times (columns), from a time_bins array, for all samples of X (rows).	required
final_time	Float	Final time for extrapolation	required
intervals	Int	Time in each interval between last time in survival dataframe and final time	required
lags	Int	Lags to compute constant risk. if negative, will use the last "lags" values if positive, will remove the first "lags" values if 0, will use all values	-1

Returns:

Type	Description
pd.DataFrame	Survival dataset with appended extrapolated windows

Source code in xgbse/extrapolation.py

def extrapolate_constant_risk(survival, final_time, intervals, lags=-1):
    """
    Extrapolate a survival curve assuming constant risk.

    Args:
        survival (pd.DataFrame): A dataframe of survival probabilities
            for all times (columns), from a time_bins array, for all samples of X (rows).

        final_time (Float): Final time for extrapolation

        intervals (Int): Time in each interval between last time in survival dataframe and final time

        lags (Int): Lags to compute constant risk.
            if negative, will use the last "lags" values
            if positive, will remove the first "lags" values
            if 0, will use all values

    Returns:
        pd.DataFrame: Survival dataset with appended extrapolated windows
    """

    last_time = survival.columns[-1]
    # creating windows for extrapolation
    # here we sum intervals in times to exclude the last time, that already is in surv dataframe and
    #  to include final time in resulting dataframe
    extrap_windows = np.arange(last_time + intervals, final_time + intervals, intervals)

    # calculating conditionals and hazard at each time window
    hazards = _get_conditional_probs_from_survival(survival)

    # calculating avg hazard for desired lags
    constant_haz = hazards.values[:, lags:].mean(axis=1).reshape(-1, 1)

    # repeat hazard for n_windows required
    constant_haz = np.tile(constant_haz, len(extrap_windows))

    constant_haz = pd.DataFrame(constant_haz, columns=extrap_windows)

    hazards = pd.concat([hazards, constant_haz], axis=1)

    return hazard_to_survival(hazards)