statsmodels.duration.hazard_regression.PHRegResults.predict#
- PHRegResults.predict(endog=None, exog=None, strata=None, offset=None, transform=True, pred_type='lhr')[source]#
Returns predicted values from the proportional hazards regression model
- Parameters:
- exogarray_like,
optional Data to use as exog in forming predictions. If not provided, the exog values from the model used to fit the data are used.
- endogarray_like,
optional 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.
- strataarray_like,
optional 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).
- offsetarray_like,
optional Offset values used to create the predicted values.
- pred_type{‘lhr’, ‘hr’, ‘surv’, ‘cumhaz’},
optional If ‘lhr’, returns log hazard ratios, if ‘hr’ returns hazard ratios, if ‘surv’ returns the survival function, if ‘cumhaz’ returns the cumulative hazard function.
- transformbool,
optional If the model was fit via a formula, whether to pass exog through the formula before forming the prediction.
- exogarray_like,
- Returns:
BunchA 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.