I'm always wondering whether anyone has utilized regime-switching models successfully in forecasting or trading.

Academia has long discussed this topic in-depth, such as using Regime Switching models for detection of abrupt market dislocation or structural changes. Popular techniques include modeling the underlying process as a Markov Process with certain distributions, and use such model to estimate the transition probability matrix.

The premise of this approach is attractive: if we can apply different approaches based on different market regimes, then the modeling process can better reflect the reality. However, like any other model, the reliability of such model is not convincing.

Personally, I've tried implementing some Regime-switch models proposed by academics and test their explanatory power for detecting market shifts (such as shift of cycles, volatility levels, etc), but the results are always disappointing.

Can any fellow friends here share your opinions? Cheers.

  • 1
    $\begingroup$ If my answer did help you it is good practice here to upvote and accept it - thank you :-) $\endgroup$ – vonjd Jul 1 '14 at 7:42
  • $\begingroup$ @Simon are you asking us to tell you how to make money? Because that is what it seems like... there is a reason most academic papers are disappointing. It is about trial and error, there is no such things as a free lunch, just bitcoin, just kidding, but what did you expect? $\endgroup$ – Ted Taylor of Life Sep 28 '17 at 20:12

Windham Capital Management is using hidden markov models for their Risk Regime Strategies.

Mark Kritzman, who is also CEO, has published an article about the general outline of the strategy (with source code so you can replicate the results!):

Regime Shifts: Implications for Dynamic Strategies (corrected August 2012) by M. Kritzman, S. Page, D. Turkington]


Regime shifts present significant challenges for investors because they cause performance to depart significantly from the ranges implied by long-term averages of means and covariances. But regime shifts also present opportunities for gain. The authors show how to apply Markov-switching models to forecast regimes in market turbulence, inflation, and economic growth. They found that a dynamic process outperformed static asset allocation in backtests, especially for investors who seek to avoid large losses.

Edit: The paper is now behind a paywall... if you find a free version pls. let me know in the comments, I will then update the answer.
Edit2: The following presentation may provide useful: http://boston.qwafafew.org/wp-content/uploads/sites/3/2017/01/Regime-Shifts_Turkington_QWAFAFEW.pdf

  • 1
    $\begingroup$ Is this a sales promotion from Windham, to sell their fund? $\endgroup$ – emcor Jun 27 '14 at 16:56
  • 2
    $\begingroup$ @emcor: No, not necessarily. Mark Kritzman is also a lecturer at MIT and has published a lot of other interesting research papers over the years. In general I think the more qualified quants are the less secretive they are about their ideas (because they develop superior new ideas all the time). $\endgroup$ – vonjd Jun 27 '14 at 17:00
  • $\begingroup$ Which returns did the fund make since 2012? $\endgroup$ – emcor Jun 27 '14 at 17:02
  • 1
    $\begingroup$ @emcor: I don't have any inside information, only what is publicly available. $\endgroup$ – vonjd Jun 27 '14 at 17:03
  • $\begingroup$ The link "Risk Regime Strategies" is dead :( $\endgroup$ – JejeBelfort Sep 27 '17 at 13:44

One of the most famous definition of Regimes and Regime Switching in Financial Markets comes from Wyckoff Cycle. Wyckoff believed that prices judged by supply and demand, go through periods of advance, accumulation, decline an distribution based on the movement of smart money.

In the quantitative world one can use state space models (ARIMA+Markov) to model such regime shifts. Here, in this paper https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3144169 I try to model regimes as Wyckoff cycles using econometric methods and analyze the market behavior around these regimes.

The results seem promising and this is an area that should be researched more from the perspective of Timeseries analysis and Financial modelling.

  • 1
    $\begingroup$ Again - please disclose that you are "the author". This answer is exactly the same as the one you gave in quant.stackexchange.com/questions/30139. If you want to be truly helpful as opposed to just advertise your papers, then it would be good to tailor the answer a bit more to the question. $\endgroup$ – LocalVolatility Nov 12 '18 at 11:28
  • $\begingroup$ Thanks @LocalVolatility for pointitng out. I edited my answer. $\endgroup$ – sonaam1234 Nov 12 '18 at 12:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.