Not as good as this band though. Still the best live band I've ever seen:
(I hope that embeds automatically, never done it before!)
Inserted Video
(I hope that embeds automatically, never done it before!)
You reap what you sow.
- | Commercial Member | Joined Apr 2013 | 4,362 Posts
What the hell........why not. We will resume normal thread coverage tomorrow 
Inserted Video
Inserted Video
Inserted Video
Inserted Video
Inserted Video
Inserted Video
Inserted Video
- | Commercial Member | Joined Apr 2013 | 4,362 Posts
Wow Tash. I'm gobsmacked. Stunning haunting music mate. Just what the Doctor ordered 
What a great planet we live in.
What a great planet we live in.
- Joined Jan 2009 | Status: Time to vacate... | 8,336 Posts
Memories caught in time but never forgotten
- Joined Jan 2009 | Status: Time to vacate... | 8,336 Posts
Memories caught in time but never forgotten
- Joined Jan 2009 | Status: Time to vacate... | 8,336 Posts
Memories caught in time but never forgotten
- | Commercial Member | Joined Apr 2013 | 4,362 Posts
During the music binge I had on the weekend I was trialing a Mean Reversion System and determining whether or not it possessed suitable metrics for inclusion in the range of automated systems I am currently testing.
The system is a slight variant to that described by Alan Clement from Helix Trader which has been very successfully employed over the long term as a long only strategy with stocks in the ASX300 and produced an excellent in sample and out of sample equity curve displayed below.
Full rules to his system were as follows:
While obviously a robust technique for his equities universe, I was testing the validity of a slight variant to his strategy for both long and short trades for my own investment universe described here. What I was specifically looking for was a mean reversion system that was uncorrelated to my current best performing automated system which is described as DAllMT Plus FUSEX ETF (what a mouthful). Broadly this 50:50 blended system comprises the Donchian Breakout suite with a weighted bias towards the medium term systems and includes the FUSEX ETF which tracks the S&P500 TR Index. The reason for the inclusion of FUSEX was due to it's very low correlation to the DAllMT suite of systems which would assist in smoothing the equity curve to produce the most powerful risk weighted returns and highest sharpe ratio of the options.
Prior to including the BB system that is the subject of this post, the best annualised sharpe ratio's we were getting over the period 1 Jan 2002 to 31 Jan 2016 were approximately 0.45 with corresponding max drawdowns of approximately 40% and CAGR of 7% after FM fees. While we were comparing favourably to the CTA's in our performance over the period, there was obviously room for improvement with the variability of returns.
.....anway, the results of the BB test (B3L) in isolation between 1 Jan 2002 to 31 Jan 2016 were as follows.
B3L Gross Profits before FM fees:
Notes. As with most mean reversion systems, they are regarded as convergent as opposed to divergent strategies and typically have a higher win rate but a lower profit factor than traditional trend following systems. They also typically have negative skew (lots of small wins with a few big losses) as opposed to positive skew (lots of small losses with a few big wins). The B3L system produced a win rate of approximately 66% with fairly even distribution of wins between long and short trades and a profit factor of 1.15: 1.
B3L Net Profits after FM fees , Scaled and Profits Reinvested
While not producing the same wicked equity curve as Alan Clement's system, what caught my eye was the almost zero correlation to the TF Index which raised my hopes that it could be a worth addition to the portfolio suite.
So below is an blend of the 3 discreet composite systems of The Donchian Suite (DAllMT), the FUSEX ETF and the B3L mean reversion system.
The addition of the B3L system has lifted the annualised sharpe ratio to 0.54, and has lifted CAGR after FM fees to almost 9% with approximately the same level of drawdown (40%).
Slowly but surely we are flattening the equity curve on this system suite.
Regards
C
The system is a slight variant to that described by Alan Clement from Helix Trader which has been very successfully employed over the long term as a long only strategy with stocks in the ASX300 and produced an excellent in sample and out of sample equity curve displayed below.
Full rules to his system were as follows:
While obviously a robust technique for his equities universe, I was testing the validity of a slight variant to his strategy for both long and short trades for my own investment universe described here. What I was specifically looking for was a mean reversion system that was uncorrelated to my current best performing automated system which is described as DAllMT Plus FUSEX ETF (what a mouthful). Broadly this 50:50 blended system comprises the Donchian Breakout suite with a weighted bias towards the medium term systems and includes the FUSEX ETF which tracks the S&P500 TR Index. The reason for the inclusion of FUSEX was due to it's very low correlation to the DAllMT suite of systems which would assist in smoothing the equity curve to produce the most powerful risk weighted returns and highest sharpe ratio of the options.
Prior to including the BB system that is the subject of this post, the best annualised sharpe ratio's we were getting over the period 1 Jan 2002 to 31 Jan 2016 were approximately 0.45 with corresponding max drawdowns of approximately 40% and CAGR of 7% after FM fees. While we were comparing favourably to the CTA's in our performance over the period, there was obviously room for improvement with the variability of returns.
.....anway, the results of the BB test (B3L) in isolation between 1 Jan 2002 to 31 Jan 2016 were as follows.
B3L Gross Profits before FM fees:
Notes. As with most mean reversion systems, they are regarded as convergent as opposed to divergent strategies and typically have a higher win rate but a lower profit factor than traditional trend following systems. They also typically have negative skew (lots of small wins with a few big losses) as opposed to positive skew (lots of small losses with a few big wins). The B3L system produced a win rate of approximately 66% with fairly even distribution of wins between long and short trades and a profit factor of 1.15: 1.
B3L Net Profits after FM fees , Scaled and Profits Reinvested
While not producing the same wicked equity curve as Alan Clement's system, what caught my eye was the almost zero correlation to the TF Index which raised my hopes that it could be a worth addition to the portfolio suite.
So below is an blend of the 3 discreet composite systems of The Donchian Suite (DAllMT), the FUSEX ETF and the B3L mean reversion system.
The addition of the B3L system has lifted the annualised sharpe ratio to 0.54, and has lifted CAGR after FM fees to almost 9% with approximately the same level of drawdown (40%).
Slowly but surely we are flattening the equity curve on this system suite.
Regards
C