ModelGroup

ModelGroup(models[, title])

A group of models that can be treated as a single model for estimation.

Attributes

Parameters

ModelGroup.pf	A DataFrame of the model parameters.
ModelGroup.pvals	An array of the current parameter values.
ModelGroup.pnames	An array of the current parameter names.
ModelGroup.pholdfast	An array indicating which parameters are marked as holdfast.
ModelGroup.pnullvals	An array of the current parameter null values.
ModelGroup.pmaximum	An array of the current parameter maximum values.
ModelGroup.pminimum	An array of the current parameter minimum values.
ModelGroup.pbounds	A copy of the current min-max bounds of the parameters.
ModelGroup.pstderr	An array of the current parameter standard errors.
ModelGroup.mixtures

Estimation Results

ModelGroup.most_recent_estimation_result	A copy of the result dict from most recent likelihood maximization.
ModelGroup.possible_overspecification	Possible overspecification of the model.

Methods

Setting Parameters

ModelGroup.set_values([values])	Set the parameter values for one or more parameters.
ModelGroup.lock_value([name, value])	Set a fixed value for a model parameter.
ModelGroup.set_cap([cap])	Set limiting values for one or more parameters.

Parameter Estimation

ModelGroup.estimate(*args, **kwargs)	Maximize loglike, and then calculate parameter covariance.
ModelGroup.maximize_loglike(*args, **kwargs)	Maximize the log likelihood.
ModelGroup.calculate_parameter_covariance([...])	Calculate the parameter covariance matrix.

Model Fitness

ModelGroup.loglike_null([use_cache])

Compute the log likelihood at null values.

Reporting

ModelGroup.parameter_summary()	Create a tabular summary of parameter values.
ModelGroup.to_xlsx(filename[, save_now, ...])	Write the estimation results to an Excel file.

Ancillary Computation

ModelGroup.total_weight()	Compute the total weight across all models in the group.
ModelGroup.loglike([x, check_if_best])	Compute the log likelihood of the group of models.
ModelGroup.logloss([x, start_case, ...])
ModelGroup.d_loglike([x, return_series])	Compute the gradient of the log likelihood of the group of models.
ModelGroup.d_logloss([x, start_case, ...])
ModelGroup.loglike_casewise([x])	Compute the log likelihood case-by-case for the group of models.

Troubleshooting

ModelGroup.doctor(**kwargs)

Run diagnostics, checking for common problems and inconsistencies.