Template-Type: ReDIF-Paper 1.0

Author-Name: Denis Koshelev
Author-Email: koshelevdm@cbr.ru
Author-Workplace-Name: Bank of Russia, Russian Federation

Author-Name: Alexey Ponomarenko
Author-Email: ponomarenkoaa@cbr.ru
Author-Workplace-Name: Bank of Russia, Russian Federation

Author-Name: Sergei Seleznev
Author-Email: seleznevsm@cbr.ru
Author-Workplace-Name: Bank of Russia, Russian Federation

Title: Amortized Neural Networks for Agent-Based Model Forecasting 

Abstract: In this paper, we propose a new procedure for unconditional and conditional forecasting in agent-based models. The proposed algorithm is based on the application of amortized neural networks and consists of two steps. The first step simulates artificial datasets from the model. In the second step, a neural network is trained to predict the future values of the variables using the history of observations. The main advantage of the proposed algorithm is its speed. This is due to the fact that, after the training procedure, it can be used to yield predictions for almost any data without additional simulations or the re-estimation of the neural network.

Length: 36 pages
Creation-Date: 2023-07
Revision-Date:
Publication-Status:
File-URL: https://cbr.ru/StaticHtml/File/149735/WP_115_e.pdf
File-Format: Application/pdf
File-Function:
Number:wps115
Classification-JEL: C11, C15, C32, C45, C53, C63.
Keywords: agent-based models, amortized simulation-based inference, Bayesian models, forecasting, neural networks.
Handle:RePEc:bkr:wpaper:wps115