Authors: Sergey Kamenshchikov
The goal of this research was applying a nonlinear approach to the detection of market regime transitions: mean reversion to momentum regimes and vice versa. It has been shown that the transition process has nonlinear scenarios: slow and fast bifurcations. Slow bifurcation assumes that control parameter is changing slowly in relation to the system characteristic time. Gradual absorption of information provides stability loss delay effect. Fast bifurcation has a discrete non equilibrium nature. Each transition from one attracting cycle to another one is preceded by passing through fixed point state – an effect of precatastophic stabilization exists. Two analytical methods have been developed for recognition of slow and fast bifurcation: R analysis and D analysis correspondingly. Combined R/D tool has been incorporated for analysis of world financial crisis of 2008. It turned out that R analysis is more convenient for long term investment while D analysis suggests middle- and short-term approach. R/D analysis has been applied as a filter for currency positional trading system. Slow and fast bifurcation patterns have been applied for the filtering of breakdown signals. Incorporation of a filter allowed to reduce twice the number of trades and to increase system efficiency, Calmar ratio, by seven times. R/D filter allowed decreasing sensitivity to volatility: duration of equity stagnation has fallen down to two months in relation to one year for the original breakdown system. It has been shown that R and D patterns may improve the long term efficiency and stability of a momentum quantitative trading model.
Comments: 16 Pages. This paper has been accepted in Quantitative Finance Journal
[v1] 2015-07-11 14:23:05
Unique-IP document downloads: 44 times
Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.