The CSSA Regime Indicator was recently presented by David Varadi in his blog. This is an interesting indicator that appears to have performed exceptionally well in a few major stock indexes. In this blog, the ability of this indicator in identifying stable and chaotic regimes is analyzed in the case of 10 major ETFs.
The rationale behind this indicator, as explained in CSSA blog, was the use of a self-similarity metric for the purpose of identifying stable and chaotic regimes. The idea is that gains in performance come mainly from stable regimes and that was confirmed in the case of S&P 500 index with data from 1963. The rules for the indicator are listed below [see references at the end]:
- Find the high minus the low for each day going back 10 days
- Take the sum of these values (sum of the pieces)
- Find the 10-day range by taking the 10-day maximum (including the highs) and subtracting out the 10-day minimum (whole range)
- Divide the sum of the pieces by the whole range- this is a basic measure of fractal dimension/complexity
- Take the 60-day average of the 10-day series of the complexity values- this is the quarterly “chaos/stability” metric
- Use either the 252-day normsdist of the z-score or the percentile ranking of the chaos/stability metric
- Values above .5 indicate that the market is in a “chaos” regime and is much less predictable and non-stationary, values below .5 indicate that the market is stable and much more predictable.
In addition to the above rules, in my analysis I used the open as the entry price because for this type of indicator it may be difficult to project closing price levels in advance that trigger positions at the close. Commission of $0.01/share was applied. The available equity was invested at each long position. Two different systems were tested:
- Stable regime: Buy when value < 0.5 and sell when value > 0.5
- Chaos regime: Buy when value > 0.5 and sell when value < 0.5
The results for ten popular ETFs are shown on the table below (EOD adjusted data from inception):
|ETF||Stable Regime CAR (%)||Chaos Regime CAR (%)||Gains come from||Buy & Hold CAR (%)|
Observations from the above table
- Most of the gains for SPY, QQQ, DIA and EEM come from the stable regime
- In the case of QQQ, DIA and XLU, the stable regime gains outperform buy and hold gains
- In the case of TLT and XLU the performance in the chaos regime is substantial
- Gains for IWM, XLE, XLF and GLD come from the chaos regime
(A) Performance is split between stable and chaos regimes in this particular study, as they are identified by the CSSA Regime Indicator. In general, gains come from both stable and chaotic regimes.
(B) For lower volatility ETFs, such as DIA and XLU, stable performance outperforms buy and hold gain. We could immediately conclude that performance in stable regime is related to low volatility but QQQ outperformance is problematic since this ETF has high mean annualized standard deviation since inception. However, after the dot com crash volatility decreased and the backtests shows that gains from the stable regime outperform buy and hold starting in 2003. Therefore, this indicator seems to perform best in low volatility markets, according to this analysis.
(C) If quantitative methods depend on low volatility, then they may perform better in stable regimes. This is to be expected. However, this does not imply that performance will be high. It could be that a substantial portion of the gains is realized in the chaotic regime.
(D) Holding period in both stable and chaotic regimes is between 25 and 50 days.
(E) This is an interesting indicator.
I have been working on a different randomness indicator since last year. I will report the findings in a future blog post.
You can subscribe here to notifications of new posts by email.
Disclosure: no relevant positions.
© 2015 Michael Harris. All Rights Reserved. We grant a revocable permission to create a hyperlink to this blog subject to certain terms and conditions. Any unauthorized copy, reproduction, distribution, publication, display, modification, or transmission of any part of this blog is strictly prohibited without prior written permission.