Australian (ASX) Stock Market Forum

Nick Radge's Weekend Trend Trader

Jump to new
Yes willoneau, I'm either very sceptical or extremely jealous. :)

Are you including commissions and at least 1-2 % slippage on buy and sell prices each way? Really need figures for at least 5 or more years for them to mean much.

I'm extremely thanks for Radge and WTT showing the way to weekly timeframe ASX trading. I'd learned lots trying to trade a modified Bollinger BO strategy (personal interpretation of BBOF from Holy Grails) for many years, but stopping during 2015 to step back and spend more time looking at weekly data/signals/Amibroker code - which was helpful in so many ways.

Interesting Nick seems to be going more and more the way of Howard Bandy - shorter timeframes (even intra-day) on US markets including mean reversion. There is always something to learn.......
 
Hi Newt , the backtesting takes into account commissions I use Commsec rates as I use them but unable to incorporate slippage as trade signals are taken from open. I have experienced slippage and not getting orders fully filled, Skate says he uses pre-open to help with that problem.
I agree with the larger time frame but am not confident that survivor bias doesn't becomes an issue when I backtest much further than three years. I still backtest way out and around important dates to see how system would have faired.
 
Last edited:
Newt said:
Yes willoneau, I'm either very sceptical or extremely jealous. :)

Are you including commissions and at least 1-2 % slippage on buy and sell prices each way? Really need figures for at least 5 or more years for them to mean much.

I'm extremely thanks for Radge and WTT showing the way to weekly timeframe ASX trading. I'd learned lots trying to trade a modified Bollinger BO strategy (personal interpretation of BBOF from Holy Grails) for many years, but stopping during 2015 to step back and spend more time looking at weekly data/signals/Amibroker code - which was helpful in so many ways.

Interesting Nick seems to be going more and more the way of Howard Bandy - shorter timeframes (even intra-day) on US markets including mean reversion. There is always something to learn.......
Click to expand...

I was recently listening to Nick talk and he definitely seems to be moving to shorter time frames. He’s been talking a fare bit about mean reversion lately and earlier this year he did explain some of the reasoning behind this, which made sence. However, he did make it clear he wasn’t yet trading intra day here or the US.
 
From what I understand the US is more suited to mean reversion and AU trend following. Something Nick stated, as I don't trade US markets.
 
Pretty sure he does trade a semi-automated daily strategy on US that is flat by close of trading - have seen him sharing results on Twitter - some big up and down days over last couple of weeks (as you'd expect).

Always find it refreshing when people are honest about how they're handling the tough days, not just the nicely profitable ones!
 
willoneau said:
Most that traded 2019 saw an amazing first six months then flat, well I did anyway after in hindsight. My system took trades all the way up until July then not again until December.
Click to expand...

I think that was the case for most people. As far as a ratio of CAR/MDD, I've been accepting something along the lines of 1.75 to 2. Not sure what other people's expectations are on that.

willoneau said:
One thing I have noticed after backtesting is that more trades doesn't always result in more profits.
Click to expand...

In my limited experience, that seems to be the case with my testing. I wonder if that would be the case on something like the US market where you are bound to have way more signals then you could take and a deeper market. Something I want to test later on.

Newt said:
I'm extremely thanks for Radge and WTT showing the way to weekly timeframe ASX trading. I'd learned lots trying to trade a modified Bollinger BO strategy (personal interpretation of BBOF from Holy Grails) for many years, but stopping during 2015 to step back and spend more time looking at weekly data/signals/Amibroker code - which was helpful in so many ways.

Interesting Nick seems to be going more and more the way of Howard Bandy - shorter timeframes (even intra-day) on US markets including mean reversion. There is always something to learn.......
Click to expand...

willoneau said:
From what I understand the US is more suited to mean reversion and AU trend following. Something Nick stated, as I don't trade US markets.
Click to expand...

What is his Bollinger BO strategy like? There has been a temptation to buy his code.

And I think mean reversion would work better on the US market. One thing would be slippage and commissions. Any mean reversion strat I have tried on ASX has failed miserably, especially when you include commissions.
 
The ranking mrchanism he is using uses the CBT to rank signals, I believe this is used both in the back test and exploration. This eliminates selection bias in the back testing.

I’ve been trading a mean reversion system on the ASX and returned in the mid 30s% last year. You pay more commissions but I just look at it as a cost of doing business. The lack of margin now with interactive brokers is the main issue
 
Roller_1 said:
The ranking mrchanism he is using uses the CBT to rank signals, I believe this is used both in the back test and exploration. This eliminates selection bias in the back testing.

I’ve been trading a mean reversion system on the ASX and returned in the mid 30s% last year. You pay more commissions but I just look at it as a cost of doing business. The lack of margin now with interactive brokers is the main issue
Click to expand...

how did your mean reversion perform during the July/Dec period of last year compared to the Jan/June period?
 
Warr87 said:
I think that was the case for most people. As far as a ratio of CAR/MDD, I've been accepting something along the lines of 1.75 to 2. Not sure what other people's expectations are on that.
Click to expand...

I'm in the process of educating myself. Can you please explain CAR/MDD ratio. The definition and how its used.
 
Quote from ASX Market Watch - Your CAR / MDD is your Compound Annual Return divided by your Maximum Draw Down. It is a good example of how quickly your system can recover after its largest draw-down. For example, if your system experienced a maximum draw down of 35%, and its average annual returns were 21%, then we use 21 divided by 35 to get 0.60. Anything over 0.50 is considered good – this would mean it would take on average around 2 years to recover from your largest draw-down. If you have a CAR / MDD greater than 1.00, this is considered excellent, as typically your system makes enough on average in a single year to compensate for your largest draw-down.
 
Roller_1 said:
I’ve been trading a mean reversion system on the ASX and returned in the mid 30s% last year. You pay more commissions but I just look at it as a cost of doing business. The lack of margin now with interactive brokers is the main issue
Click to expand...

I'd love to hear more about how you got a mean reversion system to work. Was it daily, weekly, or intra-day? I'm assuming the last 6months of last year were good for your system.

willoneau said:
Quote from ASX Market Watch - Your CAR / MDD is your Compound Annual Return divided by your Maximum Draw Down. It is a good example of how quickly your system can recover after its largest draw-down. For example, if your system experienced a maximum draw down of 35%, and its average annual returns were 21%, then we use 21 divided by 35 to get 0.60. Anything over 0.50 is considered good – this would mean it would take on average around 2 years to recover from your largest draw-down. If you have a CAR / MDD greater than 1.00, this is considered excellent, as typically your system makes enough on average in a single year to compensate for your largest draw-down.
Click to expand...

I had never really looked up what was an acceptable number. For me, I couldn't try and get 15% returns while risking 30% DD, hence why I set my requirements higher. This is also on the basis that my MDD from backtest will be higher than what I actually get. Time will tell if my expectations are too unrealistic.
 
Roller_1 said:
The ranking mrchanism he is using uses the CBT to rank signals, I believe this is used both in the back test and exploration. This eliminates selection bias in the back testing.

I’ve been trading a mean reversion system on the ASX and returned in the mid 30s% last year. You pay more commissions but I just look at it as a cost of doing business. The lack of margin now with interactive brokers is the main issue
Click to expand...

Can you clarify on CBT please?
 
Joe90 said:
Can you clarify on CBT please?
Click to expand...

Custom Back Test for AmiBroker, I am assuming that he employs expert programmers for the Low-level approach of back testing and customisation of ranking signals

Extracts from AmiBroker help file

upload_2020-2-6_18-32-47.png

upload_2020-2-6_18-30-42.png
 

Similar threads

W
Monthly Momentum - ASX300 Super
6 7 8
Replies
155
Views
36K
Warr87
W
MovingAverage
My New Monthly System
2 3 4
Replies
79
Views
20K
jon the pom
jon the pom
Top