I was wondering if anyone has any experience with regime shifts in the markets? Perhaps methods of predicting regime shifts? I’ve looked into neural networks, and a couple mathematical models (Markov) but they’re still a little out of my reach. Just wondering if anyone else has looked into it
Thanks,
Clark
You’ll have to define for us what you mean by “regime shifts”.
As defined on Wikipedia, they are “[I]large, abrupt, persistent changes in the structure and function of a system. Regimes are considered persistent relative to the time period over which the shift occurs. The change of regimes, or the shift, usually occurs when a smooth change in an internal process (feedback) or a single disturbance (external shocks) triggers a completely different system behavior.[/I]”.
So mainly these would be triggered by fundamental changes (external shocks) in the context of the financial markets. I’m looking at the weekly and monthly timeframes for theses changes. It is quite obvious when regime changes has taken place. There has been some mathematical modelling as well as machine learning done in terms of analysing these changes and perhaps predict future changes. These would involve nothing but price, volume and time rather than fundamental data.
The way I see it is if you can prepare for these regime shifts, or even identify the one we are in currently, you have a significant edge in the market, especially if you are running multiple systems.
If you want to go by the Wikipedia definition and focus on changes in the “structure and function of a system” then what you’re really talking about is the framework within which trades happens. These things tend to be more regulatory oriented. For example, the introduction of the EUR was a regime change. You could make the case that the rise of HFT in the equity market is a regime change brought on by the shifts in the regulatory environment there. The way central banks and financial authorities interact with exchange rates could be considered a regime change. From that perspective, it would take both the monitoring of regulatory type of developments and the ability to interpret the implications of those changes to anticipate regime changes.
If, however, you are simply talking about the way prices move because market participants have shifted how they think about and react to news and information then that’s a different thing.