A bootstrap test for threshold effects in a diffusion process.

Autor:	Rachinger, Heiko, Lin, Edward M. H., Tsai, Henghsiu
Předmět:	MONTE Carlo method MAXIMUM likelihood statistics STOCHASTIC differential equations
Zdroj:	Computational Statistics; Jul2024, Vol. 39 Issue 5, p2859-2872, 14p
Abstrakt:	This paper proposes a bootstrap testing approach based on an approximate maximum likelihood method to discern whether a diffusion process is linear or whether there are threshold effects in the drift, the diffusion term or in both. It complements an alternative method based on the least-squares estimator which focuses on threshold effects in the drift. Monte Carlo simulations illustrate that the proposed testing approach is able to detect the source of the non-linearity. Two empirical applications show the importance of modeling threshold effects in the diffusion instead of the drift. [ABSTRACT FROM AUTHOR]
Databáze:	Complementary Index
Externí odkaz:	Zobrazit plný text záznamu Full text from SpringerLink