Download the Jupyter Notebook for this section: integration.ipynb

# Integration Example¶

[1]:

import pyblp

pyblp.__version__

[1]:

'0.10.1'


In this example, we’ll build a Monte Carlo configuration with 1,000 draws for each market and a fixed seed.

[2]:

integration = pyblp.Integration('monte_carlo', size=1000, specification_options={'seed': 0})
integration

[2]:

Configured to construct nodes and weights with Monte Carlo simulation with options {seed: 0}.


Depending on the dimension of the integration problem, a level six sparse grid configuration may have a similar number of nodes. However, even if there are fewer nodes, it is likely to perform better in the BLP problem. Sparse grid construction is deterministic, so a seed is not needed to fix the grid every time we use this configuration.

[3]:

integration = pyblp.Integration('grid', size=7)
integration

[3]:

Configured to construct nodes and weights in a sparse grid according to the level-7 Gauss-Hermite rule with options {}.