Dump it Here

[rnr]

Hi @Skate,
Just wondering whether your "GTFO" indicator is based on a single function (e.g. say % drop in price of a stock) or is there another function which perhaps indicates an exit signal for all open trades?
Cheers,
Rob

 

[Newt]

Thanks for speaking up Trendnomics. Your moniker about Sharpe chasing is always in mind during system development. Very glad to hear you're active, and even more glad you're chiming in to challenge the discussions.

I'm glad you have fully embraced the Vulcan control of emotions lifestyle, but unfortunately I for one still suffer from some human mortal tendancies.

If I understand correctly, you think there would be benefit from a more public analysis of statistical factors underlying the momentum effect in stock prices?

(This sort of thinking?)
https://www.semanticscholar.org/pap...nden/1453464b12d66c43ba6d1b28b58baf4871f6b4bc

 

[Trendnomics]

If I understand correctly, you think there would be benefit from a more public analysis of statistical factors underlying the momentum effect in stock prices?

(This sort of thinking?)
https://www.semanticscholar.org/pap...nden/1453464b12d66c43ba6d1b28b58baf4871f6b4bc

Reading the relevant published literature can't hurt (although it can be convoluted at times). But ultimately, each TF system developer/trader has to be able to answer the most important question (i.e. What is the one statistical anomaly in the price action, which my system is exploiting to be profitable, over the long term?). Without an answer to this, you may be ill-equipped to endure the long road. Concurrently, all pilots have a fundamental understanding of aerodynamics, and will be able to answer their most important question (i.e. What keeps the airplane I am flying in the air?).

We can begin the journey of answering the most important question, with more questions:

1. For freely traded financial instruments, is the underlying price action random or predictable?
2. If random, how can my TF system possibly be profitable?
3. If predictable, why are most of my long time-frame TF system back-tests profitable, but most of my short time-frame system back-tests are non-profitable? If predictable, surely more of my system back-tests would be profitable (particularly over a variety of time-frames)?
4. Is question #1 really that black and white (for all time-frames)?

 

[qldfrog]

Well I have just PM'd Mr Skate to see if he can code it up. Actually, all but 1 should be relatively straightforward. I guess we wait and see if he can.

As to the issue of 'objectivity' (coded) and subjectivity (observation), there may be issues, hence my initial challenge to Mr Skate (to see if it is possible) to code it, taking it from discretionary to systematic.

jog on
duc

there are very very few thinks which can not be coded Duc as long as it is based on measured values/acquired data be it pooled, equation results etc even number of references in twitter etc

 

[ducati916]

there are very very few thinks which can not be coded Duc as long as it is based on measured values/acquired data be it pooled, equation results etc even number of references in twitter etc

So Mr Frog,

The initial results are in from Mr Skate. I'll let him update the thread, he may have some late improvements to add.

jog on
duc

 

[Skate]

Boom! Best post on this whole thread! I've been a long time reader of this thread, and two notable things have become apparent:
1. Most of the traders on here, still haven't FULLY accepted that systematic trend-following, will always involve losses and large draw-downs
2. Most of the traders on here, still believes that running multiple trend-following systems will provide some form of diversification or overall draw-down protection

Boom !!
Your post is brillant & your views are welcomed.

I've been a long time reader of this thread
You making the remark that you are a "long time reader of this thread" is "music to my ears".

Let me answer #1 & #2
1. Guilty as charged - large drawdowns are acceptable, whereas massive drawdowns are harder to stomach making others re-evaluate their coding ability searching for strategy improvements - In my case I'm unable to fault my current exit strategy, thus the recent frustration.
2. Guilty as charged - I run multiple trend-following systems that provide "diversification" in entering & exiting the markets only. Moving sizable amounts in & out of the markets is a challenge. Putting large bets on is a killer, as slippage is the enemy. The recent volatility & panic in the markets has allowed me to increase my bets 10 fold to take advantage of the current trading environment. I've posted that I win big as well as losing big. I do concede however that having a variety of strategies doesn't give additional drawdown protection.

Trading versus investing
My trading is not diversified whereas my overall investment strategy is.

@Trendnomics, thanks for making a valuable input in the "Dump it here" thread as there are only a few members who take the time or make the effort to post in the hope of helping others.

Skate.

 

[jjbinks]

We can begin the journey of answering the most important question, with more questions:

1. For freely traded financial instruments, is the underlying price action random or predictable?
2. If random, how can my TF system possibly be profitable?
3. If predictable, why are most of my long time-frame TF system back-tests profitable, but most of my short time-frame system back-tests are non-profitable? If predictable, surely more of my system back-tests would be profitable (particularly over a variety of time-frames)?
4. Is question #1 really that black and white (for all time-frames)?

Although I don't have vigorous mathematical proof my reading has convinced that TF system profit due to two factors
1)"Momentum" phenomena which makes price actions non-random
2)Diversification/Position sizing - Buying say 20 stocks with position size decided by risk or pct/equity rather than market cap as would be the case when buying the index is more often than not likely to be more profitable than just buying the index. (This because in most time periods small/medium caps outperform large caps over longer time frame)

 

[Skate]

So Mr Frog, The initial results are in from Mr Skate. I'll let him update the thread, he may have some late improvements to add.

@ducati916 contacted me today offering help with timing of the entry & exit using global benchmarks. The brief was to code his 5 fundamental indicators culminating into a useful code (I'll be forever grateful). The information Duc has shared is confidential.

The "Duc Indicator"
The information Duc shared was in the hope I could turn it into a useable indicator that I could use in one of my strategies. The "Duc Indicator" forms part of the buy condition as well as the sell condition. The indicator has blown me away as it works so well, maybe too well (further testing is required). At this stage I can't fault the "Duc Indicator" & I've been playing with it all day. The indicator is definitely not forward looking. I've used the "MAP Weekly Strategy" & my "BlueWren Daily Strategy" as the guinea pigs.

Lets see how the "Duc Indicator" handles a Daily Backtest
Backtest period: 1st July 2019 to last Thursday - This is a screen capture using the Daily "Duc Indicator" on my "BlueWren Daily Strategy"






The "Duc Indicator" backtest - Portfolio Equity Chart
Backtest period: 1st July 2019 to last Thursday - This is a screen capture using the Daily "Duc Indicator" on my "BlueWren Daily Strategy"







Lets see how the "Duc Indicator" handles a Weekly Backtest
Backtest period: 1st July 2019 to last Thursday - This is a screen capture using the Weekly "Duc Indicator" on my "MAP Weekly Strategy"






The backtest Weekly Portfolio Equity Chart
Backtest period: 1st July 2019 to last Thursday - This is a screen capture using the Weekly "Duc Indicator" on my "MAP Weekly Strategy"



The "Duc Indicator"
The "Duc Indicator" has sharpened the entry but more importantly the "exit" kicked in much earlier on the 28th Feb 2020 - the strategy went straight to cash, so it works.

Explanation of the Flat line
The flat line across the top of the Portfolio Equity Chart indicates the "Duc Indicator" is off. (no trades are allowed to be taken till the "Duc Indicator" turns back on)

Summary
I hope the graphic displays how well the "Duc Indicator" enters & exits positions quickly "trading a lot" is the price you have to pay for added protection I guess.

Thanks for the ideas & help Duc..

Skate.

 

[Rsthree]

Indeed apologies for using the acronym without first explanation
Phone typing is not an easy sport

Hehe, I was thinking too much, I thought SL was some exotic exit filter.

 

[Newt]

Sounds impressive Skate and Duc.
How about 2015 and late 2018 - did it help there too?

Dammit, where are all the Chinese hackers when you need them?

 

[Skate]

Sounds impressive Skate and Duc.How about 2015 and late 2018 - did it help there too? Dammit, where are all the Chinese hackers when you need them?

@Newt, all I can say is "I believed Duc was a smart bunny" - NOW I know for sure.

The "Duc Indicator" handled 2015 & 2018 on it's ear, pity I don't have data for 2007 & 2008.

2015






2018

 

[Trendnomics]

Although I don't have vigorous mathematical proof my reading has convinced that TF system profit due to two factors
1)"Momentum" phenomena which makes price actions non-random
2)Diversification/Position sizing - Buying say 20 stocks with position size decided by risk or pct/equity rather than market cap as would be the case when buying the index is more often than not likely to be more profitable than just buying the index. (This because in most time periods small/medium caps outperform large caps over longer time frame)

1) "Momentum" is a consequence of the statistical anomaly - we are currently saying, that the airplane flies due to it having wings (current answer depth).
2)Diversification/Position sizing does not attribute to a system being profitable or unprofitable - it does however contribute to the degree of profitability for a system with a positive expectancy + it determines the risk of ruin and ruin level of a system. But no amount of Diversification/Position sizing, will result in a long term profit, for a system with a negative expectancy. Diversification/Position sizing also has no impact on the underlying price action of a financial instrument. We can safely eliminate this as an answer.

 

[Newt]

Yikes, that does look very promising.

Congrats to you both if the outcomes if mutually beneficial. You derserve some reward for all the paying forward effort in this thread Skate.

Also double underlines Duc comments for all to listen to as well. Suspect only small bits of info getting through before as Duc clearly playing in markets and strategies out the usual TF ASX norm for most of us.

 

[Newt]

Ok, I'm back for tidbits again.

Would all the component data of the super Duc model be available to Norgate subscribers, hypothetically speaking, or does some macro data require hunting down from elsewhere??

 

[Skate]

Ok, I'm back for tidbits again.Would all the component data of the super Duc model be available to Norgate subscribers, hypothetically speaking, or does some macro data require hunting down from elsewhere??

Newt, Norgate data has all the required global indexes to code the “Duc Indicator”. Also, Duc has given me help expecting nothing in return. You only have to read a few of his posts to understand how knowledgeable he is.

Skate.

 

[Newt]

Yes.....
Thanks Skate.

Quite an Easter - ideas and inspiration on all fronts here.....

 

[Bazzi]

Thanks Skate!
I look forward for your buy/sell update with duc indicator reaffirmation

 

[ducati916]

Newt, Norgate data has all the required global indexes to code the “Duc Indicator”. Also, Duc has given me help expecting nothing in return. You only have to read a few of his posts to understand how knowledgeable he is.

Skate.

Well actually Mr Skate, I'm hoping you keep the updates coming so that it can be truly assessed under real market conditions going forward.

Actually Mr Skate is being unduly modest: my methodology, being discretionary, can be quite subjective. Mr Skate undertook the heavy lifting in coding and thereby adding the objectivity to my interpretation.

jog on
duc

 

[Skate]

Hi @Skate,
Just wondering whether your "GTFO" indicator is based on a single function (e.g. say % drop in price of a stock) or is there another function which perhaps indicates an exit signal for all open trades?
Cheers,
Rob

Rob, the "GTFO Indicator" is a crude indicator just as an Index Filter is. For others, a basic "Index filters" turns-on & turns-off whether the close is above or below a moving average. A "Rate of Change" (ROC indicator) is a momentum indicator. For both indicators to work efficiently the lookback period is critical & totally independent of each other. In saying this the ROC Filter & Index Filter have been optimised to activate on the same bar.

The "GTFO Indicator"
The GTFO indicator sprung into life after looking at a lot of charts & remarking "if only I exited here" or "here". I had the grand idea to backtest the idea, exiting all my positions when the Index filter turned off, that idea was quickly shelved. My next grand idea was to exit when my ROC was below 0% (when the yellow ribbon turns on) another brain-fade of an idea. My next attempt was to optimise the ROC Filter to come on at the exact same bar the Index Filter turned off (Ribbon colours, ROC below 0% = YELLOW Ribbon whereas when the Index Filter is off = RED Ribbon. When the yellow ribbon turns on & the Index Filter is Red = I quickly "GTFO". That's the GTFO filter in a nutshell. Applying those parameters to individual stocks & "wam-bam-thank-you-mam" the GTFO indicator was born.

The "Duc Indicator"
The GTFO indicator is crude & has been optimised to my Index (XAO) Filter whereas the "Duc Indicator" is very detailed & exotic indicator linked to the performance of a Global Index. When Duc gave me the brief I had to do a lot of reading (supplied by Duc) not only to understand his idea but how to go about coding it. My first thought was "how the hell am I going to do this" but as usual Duc was kind enough to fully explain his methodology. Frankly I'm amazed that he can follow such a detailed plan in his head.

My Exit strategy
I'll be playing around with the "Duc Indicator" some more today but on face value my exit strategy has now gone from a Vespa to a Ducati 916

More information on the "GTFO" indicator can be found here: https://www.aussiestockforums.com/posts/1060850/

Skate.

 

[Skate]

Well actually Mr Skate, I'm hoping you keep the updates coming so that it can be truly assessed under real market conditions going forward

Duc, I certainly will to keeping others updated - I'm planning to do a lot more testing today.

Skate.