The Run Method

The run() method runs the simulation on basis of a given specification and coefficients.

Input:

specification

- an instance of class EquationSpecification specifying variables to be used in the simulation.

coefficients

- an instance of class Coefficients that contains values of coefficients to be used in the simulation.

dataset

- a Dataset representing the whole set of observations.

index

- an index within dataset determining for which observations the prediction is to be made. If it is not given, the whole dataset is considered.

chunk_specification

- a dictionary specifying how to determine the number of chunks to run the model in (see Section 4.2.3).

data_objects

- a dictionary containing other datasets and arguments needed for computing variables.

run_config

- additional Resources for controlling the simulation run.

Algorithm: 
The algorithm is implemented in the method run_chunk() called from the parent class ChunkModel for each chunk. It invokes a computation of all variables given in the specification. Then for each submodel it creates a data matrix for values corresponding to index and invokes the run() method of the object stored in the class attribute regression (see Section 7.6.5).

Output: 
The method returns an array of the same size as index, determining the outcome of the regression for each observation included in index.