Fast Estimation of Bayesian State Space Models Using Amortized Simulation-Based Inference

Ramis Khabibullin and Sergei Seleznev
Additional contact information
Ramis Khabibullin: Independent Researcher

No wps104, Bank of Russia Working Paper Series from Bank of Russia

Abstract: This paper presents a fast algorithm for estimating hidden states of Bayesian state space models. The algorithm is a variation of amortized simulation-based inference algorithms, where numerous artificial datasets are generated at the first stage, and then a flexible model is trained to predict the variables of interest. In contrast to those proposed earlier, the procedure described in this paper makes it possible to train estimators for hidden states by concentrating only on certain characteristics of the marginal posterior distributions and introducing inductive bias. Illustrations using the examples of stochastic volatility model, nonlinear dynamic stochastic general equilibrium model and seasonal adjustment procedure with breaks in seasonality show that the algorithm has sufficient accuracy for practical use. Moreover, after pretraining, which takes several hours, finding the posterior distribution for any dataset takes from hundredths to tenths of a second.

Keywords: amortized simulation-based inference; Bayesian state space models; neural networks; seasonal adjustment; stochastic volatility; SV-DSGE. (search for similar items in EconPapers)
JEL-codes: C11 C15 C32 C45 (search for similar items in EconPapers)
Pages: 38 pages
Date: 2022-12
New Economics Papers: this item is included in nep-cmp, nep-dge and nep-ets
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cbr.ru/StaticHtml/File/142400/wp_104e.pdf (application/pdf)
Our link check indicates that this URL is bad, the error code is: 403 Forbidden

Related works:
Working Paper: Fast Estimation of Bayesian State Space Models Using Amortized Simulation-Based Inference (2022)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bkr:wpaper:wps104

Access Statistics for this paper

More papers in Bank of Russia Working Paper Series from Bank of Russia Contact information at EDIRC.
Bibliographic data for series maintained by BoR Research ( this e-mail address is bad, please contact ).