Dump it Here

[qldfrog]

Opening a new subject
One of my systems got stuck with ajm..ajm will go under receivership and basically is now worthless.any way to avoid this based on your experience? Second time it happens in my systems so 10k in smoke which basically reduces much gains to zero sum overall.
Stock filters can help but not that much.any mecanical way to avoid or reduce this risk you have found..i could play within the asx 50 realm i know....
And even with historical data, can you backtest in AB and take these into account?

 

[Skate]

The Daily VIX Strategy
The original strategy has an improved volatility exit. The daily updates of the original Daily VIX Strategy have been discontinued

Replacement
The original VIX Strategy entry criteria is unaltered - but the original exit has been replaced with a volatility exit as suggested by @ducati916

"Duc's VIX Strategy"
The updated strategy will be marked as Duc's Daily VIX Strategy & the trading results will be updated weekly after the close on Friday. (not daily)

Skate.

 

[Skate]

Thank you @Skate and @ducati916 for your continued work in the VIX Strategy. Is there any chance of us having a peek at the lists of Buy Shares, Open Positions and Sold Trades that made these new results? Would love to view them on my charts.

Skate.

 

[Skate]

Skate.

 

[Skate]

Skate.

 

[rnr]

Opening a new subject
One of my systems got stuck with ajm..ajm will go under receivership and basically is now worthless.any way to avoid this based on your experience? Second time it happens in my systems so 10k in smoke which basically reduces much gains to zero sum overall.
Stock filters can help but not that much.any mecanical way to avoid or reduce this risk you have found..i could play within the asx 50 realm i know....
And even with historical data, can you backtest in AB and take these into account?

Hi @qldfrog,

Minimum Price, Turnover and Volume requirements should exclude a number of problem issues, one would assume.

Cheers,
Rob

 

[debtfree]

Thank you @Skate, much appreciated.

 

[qldfrog]

Hi @qldfrog,

Minimum Price, Turnover and Volume requirements should exclude a number of problem issues, one would assume.

Cheers,
Rob

but no garantee and especially when you target small caps.
so should the optimal number of portfolio positions be increased when dealing with these, reducing so the exposure per code; @peter2 had some interesting foray into these more speculative shares in his thread

 

[Trav.]

Hi @qldfrog,

Minimum Price, Turnover and Volume requirements should exclude a number of problem issues, one would assume.

Cheers,
Rob

@qldfrog I think that we have all been burnt by suspensions / delisting etc and I have purposely move away from the spec end of the ASX due mainly to this issue. As @rnr points out there is one option to filter out stocks but sometimes **** happens and then you have to manage your exposure (risk) for this type of event as you highlighted.

We all read about 10 or even 100 baggers but not many people publicize there bottom draw which is pretty full by chasing these elusive bag's, especially for us mugs who have no inside knowledge...

As an example have a look at how many companies were delisted from the ASX in the last 6 months - varying reasons of course but I had no idea there were that many.

Delisted companies

www2.asx.com.au

 

[peter2]

@qldfrog and other system traders. @Skate's mentioned this before and I'll repeat it again with additional comments.
A trading system is not the same as a trading plan. A trading system can be executed as a TP and this is what has been commonly presented in many journals (threads) on ASF. The benefit is that it's quick and convenient but the downside is that this method misses fairly obvious potential problems. An example, The VIX system selected CGL after a large spike in price that triggered the system buy signal. A person using the system as a TP buys on the next open. I would never have bought CGL because the price spiked higher after a takeover offer (TO) was announced.

My TP has a selection system and then the candidates go through an additional screening process. The screening process includes a brief check on the news and would have noted the takeover offer. The TO offer caps the price and this makes the potential reward too small for the risk. Trading candidates with a poor R:R mean there's no trade. Candidates that spike +100%/d or have huge range bars also fail this requirement in the secondary screening process.

The System finds the candidates. The additional screening process checks for any commonsense (fundamental) concerns;
(i) Is the potential Risk: Reward OK?
(ii) Company activities. Knowing this helps me to assess the RR also.
(iii) Cash on hand (check last quarterly report). I don't want to trade a company that is running out of cash.
(etc) I have a few others that are relevant to my personal biases so won't include them here BUT they're on my screening checklist so I don't forget to check them.

While the secondary screening process seems like a great idea. I will mention that this will cause me to miss out on great buy signals generated by the system. Eg. I will never trade old fashion media companies. I missed great short term price swings in SWM and NEC because of my bias against them. I need to remind myself that trading short term price moves is not the same as medium term weekly trades. What I think about the companies shouldn't matter in the short term when I'm trading price momentum.

I'm prepared to discuss AJM and will do so in the AJM thread.

 

[qldfrog]

@qldfrog and other system traders. @Skate's mentioned this before and I'll repeat it again with additional comments.
A trading system is not the same as a trading plan. A trading system can be executed as a TP and this is what has been commonly presented in many journals (threads) on ASF. The benefit is that it's quick and convenient but the downside is that this method misses fairly obvious potential problems. An example, The VIX system selected CGL after a large spike in price that triggered the system buy signal. A person using the system as a TP buys on the next open. I would never have bought CGL because the price spiked higher after a takeover offer (TO) was announced.

My TP has a selection system and then the candidates go through an additional screening process. The screening process includes a brief check on the news and would have noted the takeover offer. The TO offer caps the price and this makes the potential reward too small for the risk. Trading candidates with a poor R:R mean there's no trade. Candidates that spike +100%/d or have huge range bars also fail this requirement in the secondary screening process.

The System finds the candidates. The additional screening process checks for any commonsense (fundamental) concerns;
(i) Is the potential Risk: Reward OK?
(ii) Company activities. Knowing this helps me to assess the RR also.
(iii) Cash on hand (check last quarterly report). I don't want to trade a company that is running out of cash.
(etc) I have a few others that are relevant to my personal biases so won't include them here BUT they're on my screening checklist so I don't forget to check them.

While the secondary screening process seems like a great idea. I will mention that this will cause me to miss out on great buy signals generated by the system. Eg. I will never trade old fashion media companies. I missed great short term price swings in SWM and NEC because of my bias against them. I need to remind myself that trading short term price moves is not the same as medium term weekly trades. What I think about the companies shouldn't matter in the short term when I'm trading price momentum.

I'm prepared to discuss AJM and will do so in the AJM thread.

one simple rule you quote: discarding takeover offer (TO) also means you will never profit from bidding war which are quite frequent and could reach a final price well over the initial bidder offer.not all black and white;
I get what you say about TP vs TS but as soon as you introduce discretion, be it activities, domain, even cash on hands, you introduce bias or reliance on figures or reports that may be untrue, obsolete..or for me untrustworthy.
For a listed company, i trust share price present and historic, volume and not much else to be honest so my desire for system trading

 

[Skate]

Skate.

 

[Skate]

Duc's solution worked a treat
Adding Duc's exit idea to my original strategy made a huge improvement by removing my looping exit & replacing it with the information that Duc had suggested.

Summary
"Skate's Daily DUC-VIX Strategy" with @ducati916 improved exit has turned a good strategy into an excellent strategy & is now added to the stable of my trading systems.

After the US elections
I was planning to trade the Daily VIX strategy after the US elections. Even in its original form, it's a handy strategy & "I was happy to trade it".

This will be the final weekly report
As "Skate's Daily DUC-VIX Strategy" will be trading "live" as of Monday this post will be the final update.



Skate.

 

[peter2]

@Skate Thank you again for the Daily Duc-VIX updates. I marked my charts with the VIX entries and exits and as usual, I learned something from them. ( I wonder how many others did it? )

I've never chased price. When (not if) I miss the perfect entry I let it go. I don't chase price and buy it late. I do this because a higher price makes the initial risk size much larger. If I want to chase price then my pos size model would see me buying less shares as price goes higher. This larger initial risk may not be apparent to a trader who always buys a set parcel size (ignorance is bliss or the increased risk is acceptable).

The latest Duc modified exit is a very good exit strategy. Your back test results show that. What I've seen and deduced from the latest batch of VIX results is that even a late entry can be made profitable with a good exit strategy. A good exit strategy corrects the mistakes from a poor entry strategy. It may be better to say that a good exit strategy keeps the losses low in spite of the entry method used.

I've looked at the last 40 VIX trades and IMHO the entries are late in about 25% of them. I don't mean late by one or two bars. Here's two examples of what I consider late entries into a price swing on a daily chart.



It doesn't matter what happens to price from now on. Where ever the VIX system gets out of these two trades it could have done better. IMO the Daily Duc-VIX system can do better in 25% of the trades it starts.

This observation gives me another way I can push my trading performance (stretch my mindset). I can use the spec portfolio and push myself to buy even when I've missed the perfect entry. In this way I can track the results of the trades with late entries.

 

[aus_trader]

@Skate Thank you again for the Daily Duc-VIX updates. I marked my charts with the VIX entries and exits and as usual, I learned something from them. ( I wonder how many others did it? )

I've never chased price. When (not if) I miss the perfect entry I let it go. I don't chase price and buy it late. I do this because a higher price makes the initial risk size much larger. If I want to chase price then my pos size model would see me buying less shares as price goes higher. This larger initial risk may not be apparent to a trader who always buys a set parcel size (ignorance is bliss or the increased risk is acceptable).

The latest Duc modified exit is a very good exit strategy. Your back test results show that. What I've seen and deduced from the latest batch of VIX results is that even a late entry can be made profitable with a good exit strategy. A good exit strategy corrects the mistakes from a poor entry strategy. It may be better to say that a good exit strategy keeps the losses low in spite of the entry method used.

I've looked at the last 40 VIX trades and IMHO the entries are late in about 25% of them. I don't mean late by one or two bars. Here's two examples of what I consider late entries into a price swing on a daily chart.

View attachment 114296

It doesn't matter what happens to price from now on. Where ever the VIX system gets out of these two trades it could have done better. IMO the Daily Duc-VIX system can do better in 25% of the trades it starts.

This observation gives me another way I can push my trading performance (stretch my mindset). I can use the spec portfolio and push myself to buy even when I've missed the perfect entry. In this way I can track the results of the trades with late entries.

In agreement here Pete, I usually get killed chasing price. I recently made a vow to ASF that I won't chase price, so I'll stick with that.

I will watch all this high-tech action from the sidelines where it's safe and hopefully learn something.

 

[Skate]

The latest Duc modified exit is a very good exit strategy. A good exit strategy corrects the mistakes from a poor entry strategy. It may be better to say that a good exit strategy keeps the losses low in spite of the entry method used.
I've looked at the last 40 VIX trades and IMHO the entries are late in about 25% of them.

@peter2 your observations are spot on & compared to my other (weekly) trading strategies I would have to agree. I pride myself on getting into a trend early & exiting early as well when momentum slows or stalls.

The DUC-VIX Strategy
Using the VIX as an entry condition is fraught with danger without knowing how to take advantage of a great indicator. @ducati916 suggested how he uses the indicator & on face value, I was confused to understand how his methodology could be relevant with the lookback parameters he was suggesting. Like most, I've read till my eyes have bleed & I've never read anything that the DUC passed on to me. His entry condition borders on being simply brilliant.

Late entry
The late entry is directly correlated to my parameter settings & the number of filters that I use to ultimately decide on a "confirmed" entry. The entry can be sharper & I have exhausted my bag of tricks to time the entry using alternative parameter settings. The raw entry suggested by the DUC is solid as a rock & when "true" - the ribbon turns "blue". The delayed entry comes about by the additional momentum confirmations that I use at the individual level.

Duc's recent suggestion was to use (upper & Lower) volatility band
I've read @ducati916 private messages trying to glean his instructions more clearly to garner the "intent of his words". The original help was how to use the VIX indicator to its best advantage & his methodology is a beauty. The most recent help was to sharpen the entry & exit using volatility bands. With this new information, it sent me on a path of exploring ways to take advantage of his most recent suggestion.

I'm sure Duc won't mind if I quote one of his passages
"The issue is to 'predict' (or best guess the VIX) when it is topping, when it is bottoming and to steal a march on it"

My mind was racing - how to take advantage of the bands
My initial reaction was to use the upper band to take profits. The alternative - when the close touches the bottom band "exit". The theory was clean & simple till it was tested.

Take Profit Stop (not my best idea)
Using the upper Volatility Band for taking profits stifled the strategy performance so the idea was given the flick quickly. Concentrating on the exit from this point resulted in the strategy profitability improving - How? - by not exiting good positions prematurely. (in simple terms - not using a momentum exit)

More to follow
After re-reading @ducati916 "PM's" & @peter2 recent post - there is a solution I've found that is worthy of another post. The solution revolves around utilising Duc's upper volatility band & my method of implementing it.

Skate.

 

[Skate]

The DUC-VIX Strategy goes live on Monday
After exploring ways to take advantage of Duc's new suggestion - "utilising the upper volatility band" - it's too late to implement the suggestion into my VIX trading strategy before the system goes live on Monday.

Every new idea needs to be "fully tested & investigated"
I'm not sure how others test new theories but I try to "break" the advantages of any new idea. Why? because trading is a constant trade-off between risk & reward.

My first idea was a dud
Using the upper volatility band as a "Take Profit Stop" was a DUD. My latest idea - using the upper band as an additional "confirmation of volatility" at the individual level works quite well.

There are profitability improvements
Using the upper band & incorporating volatility as an additional entry condition has the benefit of lowering stock turns & reducing the false entries. The commission cost is also lower.

The upper volatility band (used as a confirmation band)
How Duc utilises the VIX as an indicator is simply brilliant "but" difficult to "get your head around at first" (well it was for me). Using the upper & lower volatility band is just another layer taking the strategy to a new level.

I've tested the idea
Using the upper band as confirmation of volatility at the individual level has pleasing results, worthy of further investigation.

The backtest results look impressive
@ducati916 raw idea of using volatility bands for entry & exits is impressive. Using the upper volatility bands to complement the entry condition - tests okay. Frankly, the improvements were immediate (without any optimisation).

Amibroker Backtest results - speaks volume
1. The backtest period is from 1st January 2020 to 6th November 2020
2. For simplicity $100k Portfolio (20 X $5k positions)
3. The only difference between the two backtest results is the backtest on the "left" has an additional entry condition using the "UPPER" volatility band to formulate a new entry. Using the upper band adds confirmation of volatility at the individual level.



Summary
I'm over the "moon" incorporating DUC's new exit strategy but slightly disappointed that there is a lack of time to truly evaluate the new volatility entry condition - "using the upper volatility confirmation band". To be honest, I wouldn't trade the new idea till it was fully tested.

Skate.

 

[Newt]

I'm fascintated by the interplay of discretionary trading and had won experience (Duc) fed into cold hard maths and algorithm (Skate). It may well be that the agorithmic interpretation of Duc's methodology misses (or averages out) the subtlety of some entries and exits.

However, for most of us, human psychology issues (fear, FOMO, over-riding signals) has the potential to damage a strong discretionary strategy to the extent I'd certainly prefer to be algorithmic every time. Greatly appreciate the tantalizing glimses of the interplay in the background between skate and duc.

On thje late entry concerns Peter2 has raised, I've also found it hard, over many years and strategies, to not incorporate momentum checks on buy signals. Yes, it does sometimes greatly delay entry and you lose profit and increase risk measured from price bases (or bottoms). However the backtesting metrics don't lie, and the sweet spot for reliability of returns versus established momentum seems to generally be skewed toward including some degree of "waiting for momentum confirmation". The average win/loss and DD figures are the metrics that improve, along with final profits. The real challenge is of course finding a way to systematically enter safely and hopefully many bars before other trend traders. This is just entries - we've heard here consistently the money is in the exit - so that is another challenge again.

 

[ducati916]

Looking first at the data generated from all the tests, a number of axioms can be stated. However, before getting to that, a look at the comment below.

This thread has taken an interesting direction with @peter2 contribution:

I've never chased price. When (not if) I miss the perfect entry I let it go. I don't chase price and buy it late. I do this because a higher price makes the initial risk size much larger. If I want to chase price then my pos size model would see me buying less shares as price goes higher. This larger initial risk may not be apparent to a trader who always buys a set parcel size (ignorance is bliss or the increased risk is acceptable).

Given that this is a 'mechanical' based thread, also quite relevant. So as an example of a mechanical system that does buy (chase price higher) higher prices, increasing the position size, we have the original Turtle System. Chunk units, which consisted of a number of futures contracts, would be bought on an increasing basis as the trend proved itself.

@Newt seems to echo the Turtles:

On the late entry concerns Peter2 has raised, I've also found it hard, over many years and strategies, to not incorporate momentum checks on buy signals. Yes, it does sometimes greatly delay entry and you lose profit and increase risk measured from price bases (or bottoms). However the backtesting metrics don't lie, and the sweet spot for reliability of returns versus established momentum seems to generally be skewed toward including some degree of "waiting for momentum confirmation". The average win/loss and DD figures are the metrics that improve, along with final profits. The real challenge is of course finding a way to systematically enter safely and hopefully many bars before other trend traders. This is just entries - we've heard here consistently the money is in the exit - so that is another challenge again.

Duc's Axiom 1

A (great) trader's purpose in trading, is to to do those things that create winning trades, while avoiding those that create losses.

Corollary: A trader's job is to create winning trades.

Axiom 2

Winning trades result from trades taken (as do losing trades).

Axiom 3

All winning trades and losing trades occur within the context of total capital.

Corollary: There is a predictable relationship between the number of winning trades made and the number of losing trades made on total capital.

Question: Do winning and losing trades represent a pure accomplishment or are they coloured by market conditions (for example bull/bear market, nominal vol. levels, news, macro-considerations, etc). If they are so coloured, how should that be accounted for?

jog on
duc

 

[ducati916]

In relation to the posited question: this equation considers (a question raised through the years on this forum) whether the ASX follows or is influenced by the US market.

My position is as below:



Now there are all manner of other effects, illusions and interesting stuff that happens. One which has been discussed on numerous occasions is 'Seasonality', which if you are in any doubt is a real phenomenon.

jog on
duc