Looking forward to your thoughts Frog as someone who sold a system using this type of indicator (modified from the known future leak of the AB zigzag indicator) was removed from purchase. Although saying that, I've no idea if it was removed because of zigzag directly or some other reason.
Dump it Here
- Thread starter Skate
- Start date
Looking forward to your thoughts Frog as someone who sold a system using this type of indicator (modified from the known future leak of the AB zigzag indicator) was removed from purchase. Although saying that, I've no idea if it was removed because of zigzag directly or some other reason.
- Joined
- 8 June 2008
- Posts
- 12,784
- Reactions
- 18,692
When i personally tried using a zz indicator based system of my ownyears ago, i was well aware of the looking forward issue and so worked on it and was expected to have solved it..but my short trial was a failure, after 3 runs aka 3 weeks, some signals diseappear new ones popped up in the past month..you give some room for human error .maybe i mistyped or was confused but issue kept repeating
that was a while back but even if not obvious or detected by ab check , there was still a future leak
I will give GB code a fair go as i am obviously interested
I have not been able to spend dev time for weeks as far too busy it will be done but be patient please
Cheers
- Joined
- 20 November 2005
- Posts
- 787
- Reactions
- 92
Glad to see you're still rabbiting on about it...what's it been? 10+ years?
FYI, my code does not have a future leak. All strategies from Unholy Grails were removed from the site several years ago, mainly for liquidity reasons. Below is an updated equity chart of the 20% Flipper.
And as a goodwill gesture, here is how you code the ZZ (20% Flipper) without a future leak
//-------------
function flipper_func(trail)
{
trailArray[ 0 ] = C[ 0 ];
for( i = 1; i < BarCount; i++ )
{
prev = (round(100*(trailArray[ i - 1 ])))/100;
if (C[ i ] > prev AND C[ i - 1 ] > prev)
{
trailArray[ i ] = Max(prev,0.8*H[ i ]);
}
else if (C[ i ] < prev AND C[ i - 1 ] < prev)
{
trailArray[ i ] = Min(prev,1.2*L[ i ]);
}
else if (C[ i ] == prev AND C[ i - 1] < prev)
{
trailArray[ i ] = 0.8*H;
}
else if (C[ i ] == prev AND C[ i - 1 ] > prev)
{
trailArray[ i ] = 1.2*L[ i ];
}
else if (C[ i ] < prev AND C[ i] > 1.2*L [ i ])
{
trailArray[ i ] = 0.8*H;
}
else if (C[ i ] > prev AND C[ i] < 0.8*H [ i ])
{
trailArray[ i ] = 1.2*L;
}
else if (C[ i ] < prev AND C[ i - 1 ] > prev)
{
trailArray[ i ] = 1.2*L[ i ];
}
else if (C[ i ] > prev)
{
trailArray[ i ] = 0.8*H[ i ];
}
else if (C[ i ] < prev)
{
trailArray[ i ] = 1.2*L[ i ];
}
else
{
trailArray[ i ] = 1.2*L[ i ];
}
}
return trailArray;
}
trail = 1;
trailArray = flipper_func(trail);
Attachments
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
@Nick Radge, thanks for supplying the updated equity chart for the 20% Flipper.
The Weekend Trend Trader
The (WTT) Turnkey Strategy backtest results I've posted below have not been optimized or "Radge Tweaks" applied. The backtest results are based on the original book settings.
My question is
Would you be kind enough to update the backtest results for the 2022 calendar year (up until today) for those interested in your "Turnkey Strategy"
Skate.
Must be close to it!
Remembered it was fixed with the above and think code by Gringotts Bank using zig() has a leak but interested if Frog finds it too
Cheers.
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
FYI
I've just asked Nick to update his WTT backtest for this calendar year.
I for one would be interested in his reply.
Skate.
I've just asked Nick to update his WTT backtest for this calendar year.
I for one would be interested in his reply.
Skate.
- Joined
- 30 June 2007
- Posts
- 7,200
- Reactions
- 1,226
Well isn't that interesting, a ZZ code with no future leak. Skate spent 10 posts arguing that it was impossible, too good to be true, danger!/beware! And yet it can be used, as I said.
Test mine if you would.
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
- Joined
- 25 July 2016
- Posts
- 322
- Reactions
- 638
I remember now why I gave up on these forums years ago...back to real life
Haha pretty much real life but face to face makes things so much easier.
Just takes a bit more time
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
Many people don't post because they don't feel like they have anything to contribute.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
This system has no future leaks AND i have tradeDelays implicitly set !!!!
AND i didn't use that nasty ZIGZAG indicator, which automatically brings up future leaks regardless of settings.
The optimum period for this system is 1 day.
NOTE: This system has 2 serious issues.
1 - 7% Drawdown.
2 - 22.13% losing trades.
For a mere bitcoin, I'll send you an optimised system where i eliminate BOTH of those 2 MAJOR flaws.
AND i didn't use that nasty ZIGZAG indicator, which automatically brings up future leaks regardless of settings.
The optimum period for this system is 1 day.
NOTE: This system has 2 serious issues.
1 - 7% Drawdown.
2 - 22.13% losing trades.
For a mere bitcoin, I'll send you an optimised system where i eliminate BOTH of those 2 MAJOR flaws.
Ok .... i cheated, but so does ZIGZAG.
Amibroker, MetaStock and others have it in their core set of indicators to use (and abuse). I'm only talking about these ones, not other implementations.
ZIGZAG Locates the exact pivot points by either looking ahead or looking back, but also changes previous pivots based on new data. Always has and always will and It's probably the only indicator that does this. It's used in many systems like wave counting and it does it's job perfectly.
Both AB and MS have strong caveats about using it.
https://www.metastock.com/customer/resources/taaz/?p=127
http://www.amibroker.com/guide/afl/zig.html
Amibroker, MetaStock and others have it in their core set of indicators to use (and abuse). I'm only talking about these ones, not other implementations.
ZIGZAG Locates the exact pivot points by either looking ahead or looking back, but also changes previous pivots based on new data. Always has and always will and It's probably the only indicator that does this. It's used in many systems like wave counting and it does it's job perfectly.
Both AB and MS have strong caveats about using it.
https://www.metastock.com/customer/resources/taaz/?p=127
http://www.amibroker.com/guide/afl/zig.html
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
@DaveDaGr8, I hope they listen to you
The dynamic nature of the ZZ indicator means you will get unrealistic results when backtesting
Amibroker's knowledge base makes a solid statement about using the Zig-Zag in trading systems. Using functions (Peak/Trough, PeakBars, Troughbars), inherently look into the future, as such they should not be used in trading system formulas.
Skate.
- Joined
- 20 November 2005
- Posts
- 787
- Reactions
- 92
- Joined
- 30 June 2007
- Posts
- 7,200
- Reactions
- 1,226
You should test my code, rather than quote what some website says. Enter the argument fairly and honestly. So far, I am the only one who has done that.
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
Don't use the Amibroker ZigZag function for backtesting or trading
It's important to remember when backtesting a strategy great care should be taken to exclude any use of revisable parameters in your strategy ensuring the backtest results are valid & accurate. I've been consistent in my messaging that "revisable parameters" inherently look into the future, & as such, should not be used
Anything is possible
But the desired results may not be.
Skate.
It's important to remember when backtesting a strategy great care should be taken to exclude any use of revisable parameters in your strategy ensuring the backtest results are valid & accurate. I've been consistent in my messaging that "revisable parameters" inherently look into the future, & as such, should not be used
Anything is possible
But the desired results may not be.
Skate.
- Joined
- 28 December 2013
- Posts
- 6,362
- Reactions
- 24,248
I couldn't have expressed it any better (From the Metastock site)
"The Zig Zag indicator filters out changes in an underlying plot (e.g., a security's price or another indicator) that are less than a specified amount. The Zig Zag indicator only shows significant changes. The Zig Zag indicator is used primarily to help you see changes by punctuating the most significant reversals.
The last leg
"It is very important to understand that the last "leg" displayed in a Zig Zag chart can change based on changes in the underlying plot (e.g., prices). This is the only indicator in this book where a change in the security's price can change a previous value of the indicator. Since the Zig Zag indicator can adjust its values based on subsequent changes in the underlying plot, it has "perfect hindsight" into what prices have done. Please don't try to create a trading system based on the Zig Zag indicator"
In a nutshell
"Its hindsight is much better than its foresight!"
Skate.
"The Zig Zag indicator filters out changes in an underlying plot (e.g., a security's price or another indicator) that are less than a specified amount. The Zig Zag indicator only shows significant changes. The Zig Zag indicator is used primarily to help you see changes by punctuating the most significant reversals.
The last leg
"It is very important to understand that the last "leg" displayed in a Zig Zag chart can change based on changes in the underlying plot (e.g., prices). This is the only indicator in this book where a change in the security's price can change a previous value of the indicator. Since the Zig Zag indicator can adjust its values based on subsequent changes in the underlying plot, it has "perfect hindsight" into what prices have done. Please don't try to create a trading system based on the Zig Zag indicator"
In a nutshell
"Its hindsight is much better than its foresight!"
Skate.
Similar threads
- Poll
