Details
- Time: 10:00-11:00AM
- Date: Friday, December 22, 2023
- Venue: MB441
- Speaker: Dr. Taosong Deng, Hunan University
- Host: Dr. Yuhao Li
Abstract
There is an increasing demand for filters applicable to nonlinear dynamic stochastic general equilibrium (DSGE) models. One of the popular choices is the particle filter. However, it is well known that this filter requires adding measurement errors to the measurement equation for its implementation. Otherwise, singularity arises, and consequently, the algorithm breaks down. Measurement errors are not a common feature of DSGE models. Motivated by this limitation, in this paper we develop a new particle filter by mapping the state vector into two subvectors: a subvector whose components are observed and a subvector whose components are latent. By only sampling and propagating particles of the latent variables, we avoid the need to introduce measurement errors. For implementation, we propose to approximate the observables’ density conditional on the latent variables using series expansions. As an important feature, the new filter also allows us to study singular DSGE models using the composite likelihood, therefore providing a unified treatment of both singular and nonlinear DSGE models.
Speaker
Dr. Taosong Deng graduated with a Bachelor's degree in Economics from Hunan University in 2012, and received a Doctorate in Economics from Boston University in 2021 under the supervision of Professor Pierre Perron.
Since 2021, he has been an Assistant Professor at the School of Finance and Statistics at Hunan University. His main research directions include theoretical and applied econometrics, quantitative macroeconomics, and empirical finance.