Comparison of recursive parameter estimation and non-linear
filtration

D 2012

Comparison of recursive parameter estimation and non-linear filtration

ČAPEK, Jan

Basic information

Original name

Comparison of recursive parameter estimation and non-linear filtration

Authors

ČAPEK, Jan (203 Czech Republic, guarantor, belonging to the institution)

Edition

1. vyd. Karviná, Proceedings of 30th International Conference Mathematical Methods in Economics, p. 85-90, 6 pp. 2012

Publisher

Silesian University, School of Business Administration

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

50200 5.2 Economics and Business

Country of publisher

Czech Republic

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

printed version "print"

RIV identification code

RIV/00216224:14560/12:00060981

Organization unit

Faculty of Economics and Administration

ISBN

978-80-7248-779-0

UT WoS

000316715900015

Keywords in English

Unscented Particle Filter;recursive analysis;time-varying parameter;DSGE model;impulse response function

Abstract

V originále

The contribution compares and contrasts the results of time-varying parameters estimation pursued by two methodologically different approaches, all in the context of a DSGE model. The original results Capek 2011 were obtained by the so-called recursive, rolling and ``first observation'' analysis. The novel methodology uses Unscented Particle Filter to filter trajectories of time-varying parameters. The comparisons are presented on the case of three parameters of a DSGE model, which demonstrate a match in the evolution of the parameter estimates. Recursive impulse response functions are compared on two cases, which also match. Although not all of the results obtained from two different methodologies completely match, the most important parameter changes are similarly captured by both approaches.

Files attached

capekREV.pdf

Citovat

ČAPEK, Jan. Comparison of recursive parameter estimation and non-linear filtration. In Ramík, J., Stavárek, D. Proceedings of 30th International Conference Mathematical Methods in Economics. 1st ed. Karviná: Silesian University, School of Business Administration, 2012, p. 85-90. ISBN 978-80-7248-779-0.

@inproceedings{992904,
   author = {Čapek, Jan},
   address = {Karviná},
   booktitle = {Proceedings of 30th International Conference Mathematical Methods in Economics},
   edition = {1},
   editor = {Ramík, J., Stavárek, D.},
   keywords = {Unscented Particle Filter;recursive analysis;time-varying parameter;DSGE model;impulse response function},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Karviná},
   isbn = {978-80-7248-779-0},
   pages = {85-90},
   publisher = {Silesian University, School of Business Administration},
   title = {Comparison of recursive parameter estimation and non-linear filtration},
   year = {2012}
}

TY  - CONF
ID  - 992904
AU  - Čapek, Jan
PY  - 2012
TI  - Comparison of recursive parameter estimation and non-linear filtration
PB  - Silesian University, School of Business Administration
CY  - Karviná
SN  - 9788072487790
KW  - Unscented Particle Filter;recursive analysis;time-varying parameter;DSGE model;impulse response function
N2  - The contribution compares and contrasts the results of time-varying parameters estimation pursued by two methodologically different approaches, all in the context of a DSGE model. The original results Capek 2011 were obtained by the so-called recursive, rolling and ``first observation'' analysis. The novel methodology uses Unscented Particle Filter to filter trajectories of time-varying parameters. The comparisons are presented on the case of three parameters of a DSGE model, which demonstrate a match in the evolution of the parameter estimates. Recursive impulse response functions are compared on two cases, which also match. Although not all of the results obtained from two different methodologies completely match, the most important parameter changes are similarly captured by both approaches.
ER  -

ČAPEK, Jan. Comparison of recursive parameter estimation and non-linear filtration. In Ramík, J., Stavárek, D. \textit{Proceedings of 30th International Conference Mathematical Methods in Economics}. 1st ed. Karviná: Silesian University, School of Business Administration, 2012, p.~85-90. ISBN~978-80-7248-779-0.

Detailed Information on Publication Record