Predictable non-linearities in U.S. inflation

Binner, J, Elgar, T, Nilsson, B and Tepper, J ORCID: https://orcid.org/0000-0001-7339-0132, 2006. Predictable non-linearities in U.S. inflation. Economics Letters, 93 (3), pp. 323-328. ISSN 0165-1765

Preview

Text 196114_157 Tepper Postprint Converted.pdf Download (96kB) | Preview

Abstract

We expand Nakamura’s (2005) neural network based inflation forecasting experiment to an alternative non-linear model; a Markov switching autoregressive (MS-AR) model. The two non-linear models perform approximately on par and outperform the linear autoregressive model on short forecast horizons of one and two quarters. Furthermore, the MS-AR model is the best performer on longer horizons of three and four quarters.

Item Type:	Journal article
Publication Title:	Economics Letters
Creators:	Binner, J., Elgar, T., Nilsson, B. and Tepper, J.
Publisher:	Elsevier
Date:	2006
Volume:	93
Number:	3
ISSN:	0165-1765
Identifiers:	Number Type 10.1016/j.econlet.2006.06.001 DOI
Rights:	© 2006 Elsevier
Divisions:	Schools > School of Science and Technology
Record created by:	EPrints Services
Date Added:	09 Oct 2015 10:13
Last Modified:	04 Feb 2022 12:25
URI:	https://irep.ntu.ac.uk/id/eprint/9698

Actions (login required)

Edit View

Statistics

Views

Downloads