statsmodels.duration.hazard_regression.PHReg.predict

PHReg.predict(params, exog=None, cov_params=None, endog=None, strata=None, offset=None, pred_type='lhr')[source]

Returns predicted values from the proportional hazards regression model.

Parameters:
  • params (array-like) – The proportional hazards model parameters.
  • exog (array-like) – Data to use as exog in forming predictions. If not provided, the exog values from the model used to fit the data are used.
  • cov_params (array-like) – The covariance matrix of the estimated params vector, used to obtain prediction errors if pred_type=’lhr’, otherwise optional.
  • endog (array-like) – Duration (time) values at which the predictions are made. Only used if pred_type is either ‘cumhaz’ or ‘surv’. If using model exog, defaults to model endog (time), but may be provided explicitly to make predictions at alternative times.
  • strata (array-like) – A vector of stratum values used to form the predictions. Not used (may be ‘None’) if pred_type is ‘lhr’ or ‘hr’. If exog is None, the model stratum values are used. If exog is not None and pred_type is ‘surv’ or ‘cumhaz’, stratum values must be provided (unless there is only one stratum).
  • offset (array-like) – Offset values used to create the predicted values.
  • pred_type (string) – If ‘lhr’, returns log hazard ratios, if ‘hr’ returns hazard ratios, if ‘surv’ returns the survival function, if ‘cumhaz’ returns the cumulative hazard function.
Returns:

  • A bunch containing two fields (predicted_values and)
  • standard_errors.

Notes

Standard errors are only returned when predicting the log hazard ratio (pred_type is ‘lhr’).

Types surv and cumhaz require estimation of the cumulative hazard function.