Optimal N-state endogenous Markov-switching model for currency liquidity timing
Wang, L. & Urga, G. ORCID: 0000-0002-6742-7370 (2025).
Optimal N-state endogenous Markov-switching model for currency liquidity timing.
Journal of Economic Dynamics and Control,
Abstract
In this paper, we examine whether globally-diversified funds' actively adjust their currency exposure in response to systematic currency liquidity movements, a behavior we term currency liquidity timing. A novel currency-liquidity-timing model embedded with an N-state endogenous Markov-switching mechanism is proposed to capture the dynamics in funds' timing behavior, as well as the external and internal drivers influencing such dynamics. Using a sample of 382 international fixed income mutual funds from July 2001 to December 2020, we find evidence of currency liquidity timing at the aggregate level for the sample funds. Interestingly, funds' currency-liquidity-timing behavior exhibits a state-switching pattern across different market periods: funds on average engage in perverse currency liquidity timing during tranquil market periods, but in positive currency liquidity timing with a stronger degree of aggressivity during more turbulent market periods. Our results suggest that the state transitions in funds' currency-liquidity-timing behavior are driven by deteriorating external currency market liquidity conditions and negative shocks to internal fund returns.
Publication Type: | Article |
---|---|
Additional Information: | © 2025. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Publisher Keywords: | Currency Factors, Endogenous Markov-Switching Models, Globally-Diversified Funds, Liquidity Timing |
Subjects: | H Social Sciences > HG Finance |
Departments: | Bayes Business School Bayes Business School > Finance |
SWORD Depositor: |
![[thumbnail of JEDC_20250201_(Manuscript&Appendices).pdf]](https://openaccess.city.ac.uk/style/images/fileicons/text.png)
This document is not freely accessible due to copyright restrictions.
Available under License Creative Commons Attribution Non-commercial No Derivatives.
To request a copy, please use the button below.
Request a copyExport
Downloads
Downloads per month over past year