Data for: Chapter 2 of "Estimation of DSGE models: Skewness matters" : Comparison of optimizers and samplers in Dynare

Guljanov, Gaygysyz; Mutschler, Willi

Other scientific publication

Abstract

This dataset contains simulated data series, results of various optimization routines, and MCMC chains. We prepared it in the course of writing "Comparison of samplers and optimizers in Dynare and in the context of Bayesian identification strength", Chapter 2 of the Doctoral thesis "Estimation of DSGE models: Skewness matters" (2024) by Gaygysyz Guljanov. We will publish the computer codes that we used for generating this dataset in a GitHub repository "https://github.com/gguljanov/comparison-optimizers-samplers-thesis". This dataset contains only raw Dynare outputs and the GitHub repository will contain Python codes for analyzing them by collecting useful pieces of information into a Pandas DataFrame. For more information about how and why we generated this dataset, we refer the readers to the above dissertation. For more information on the structure of this dataset, such as what each folder contains, we refer the readers to the README file of the above GitHub repository. Calculations (or parts of them) for this publication were performed on the HPC cluster PALMA II of the University of Münster, subsidized by the DFG (INST 211/667-1).

Details about the publication

StatusPublished
Release year2025 (30/01/2025)
Language in which the publication is writtenEnglish
DOI10.17879/44918563945
Link to the full texthttps://doi.org/10.17879/44918563945
KeywordsNumerical optimization; MCMC sampling; Metropolis-Hastings; Tailored-Randomized-Block-Metropolis-Hastings; Slice; DSGE; Weak identification; Bayesian learning-rate indicator

Authors from the University of Münster

Guljanov, Gaygysyz
Professur für VWL, Ökonometrie/Wirtschaftsstatistik (Prof. Trede)