Initialization

The class is initialized by passing the following arguments:

choice_set

- A list, array or dataset that represents the finite set of possible choices. It should have numeric values larger than zero.

utilities

- A fully qualified name of the module for computing utilities (see Section 7.6.1). Default value is 'opus_core.linear_utilities'.

probabilities

- A fully qualified name of the module for computing probabilities (see Section 7.6.2). Default value is 'opus_core.mnl_probabilities'.

choices

- A fully qualified name of the module for determining final choices (see Section 7.6.3). Default value is 'opus_core.random_choices'.

submodel_string

- If model contains submodels, this character string specifies what agent's attribute determines those submodels.

choice_attribute_name

- Name of the attribute that identifies the choices. This argument is only relevant if choice_set is not an instance of Dataset. Otherwise the choices are identified by the unique identifier of the choice_set. Default value is 'choice_id'

interaction_pkg

- This argument is only relevant if there is an implementation of an interaction dataset that corresponds to interaction between agents and choices. Default value is 'opus_core'.

run_config

- A collection of additional arguments that control a simulation run. It should be of class Resources.

estimate_config

- A collection of additional arguments that control an estimation run. It should be of class Resources.

The initialization method creates a class attribute upc_sequence, using the passed arguments utilities, probabilities and choices. It is an object of class upc_sequence (see Section 7.6.4). A class attribute choice_set is an object of Dataset. If the argument choice_set is a list or array, a Dataset is created using the values of the argument as the unique identifier called choice_attribute_name.