Dump it Here

[qldfrog]

Hi Saqeeb,

Just having a look at where you are at with paper trading your MAP system and I came across the above comment.

At the time of your post you had 18 open positions (out of 20) and no funds available to enter trades for the 2 buy signals.

Hopefully, you immediately started reviewing your system code to see where the error lies!

When you are ready to put real $$$'s at risk you must be in a position where you "own the system (including the code)" so there can be no surprises that will impact on your live trading.

Cheers,
Rob

It is an interesting case
You have 20k for 20 positions, 1k each great
after a few months, you have 15 positions
31k overall value, 2k in cash, with your 15 positions at last close.

Your explore find 4 buys...what do you do?
Do you
1) divide cash by remaining positions to fill to get the next parcel value?
2) keep to 1k per position until no cash
3)divide overall value by remaining positions to fill to get the next parcel value?and buy until cash runs out
I usually try 3...

 

[Saqeeb]

Hi Saqeeb,

Just having a look at where you are at with paper trading your MAP system and I came across the above comment.

At the time of your post you had 18 open positions (out of 20) and no funds available to enter trades for the 2 buy signals.

Hopefully, you immediately started reviewing your system code to see where the error lies!

When you are ready to put real $$$'s at risk you must be in a position where you "own the system (including the code)" so there can be no surprises that will impact on your live trading.

Cheers,
Rob

@rnr, Thank you for following my paper trading and your concern. My account was in a drawdown of close to $3,700 when my system restarted producing signals after the March sell off. As I am not rebalancing my portfolio for this paper trading experiment and my position size is $1500 each position, the portfolio at this stage is fully invested at 18 positions. In fact I should have only taken 17 positions as I would have been short of funds for the 18th position.

 

[rnr]

It is an interesting case
You have 20k for 20 positions, 1k each great
after a few months, you have 15 positions
31k overall value, 2k in cash, with your 15 positions at last close.

Your explore find 4 buys...what do you do?
Do you
1) divide cash by remaining positions to fill to get the next parcel value?
2) keep to 1k per position until no cash
3)divide overall value by remaining positions to fill to get the next parcel value?and buy until cash runs out
I usually try 3...

The answer is really simple as it lies within your system which would / should be coded accordingly.
They are your rules so you should know the answer.
As an example if you have a rule that the maximum trade size (including brokerage on acquisition) is $2500" then any trades exceeding that amount are quite simply breaching that rule.
For example, in any of Skate's systems posted on the forum there is usually a comment as when re-balancing occurs.

Cheers,
Rob

 

[qldfrog]

The answer is really simple as it lies within your system which would / should be coded accordingly.
They are your rules so you should know the answer.
As an example if you have a rule that the maximum trade size (including brokerage on acquisition) is $2500" then any trades exceeding that amount are quite simply breaching that rule.
For example, in any of Skate's systems posted on the forum there is usually a comment as when re-balancing occurs.

Cheers,
Rob

it was theoretical, I use option 3, but this highlights the fact that this implementation requires a potential buffer of $ as you might (not to say often will and must) have winners still in your portfolio with unrealized gains

 

[qldfrog]

An area which interest me and maybe others:
Mr Skate, have you consider using some of your systems in the US
There is considerable work and i believe..with no proof, that you would need to not only change parameters but also work in niche realms.
Yet, at what i believe is a Reset time, could be diversification and protection.
Have you or anyone here tried trending, or other systems in the US market.
With less emphasis on dividends, it should even be a better market for trending systems

 

[Skate]

Mr Skate, have you consider using some of your systems in the US

@qldfrog, I haven't considered trading the US markets but was curious at one stage if my Hybrid System would be tradable in other markets. It was suggested, a good system is a good system when it performs over a multitude of indexes & markets. As I was planning to invest a sizable amount with the Hybrid Strategy knowing the answer would add another layer of confidence.

I have also answered a similar question here
Short answer the Hybrid Strategy performs well in the US markets. https://www.aussiestockforums.com/posts/1057390/

What’s happening to the markets.
If you read @bigdog "NYSE-dow-jones-finished-today" thread each day it has an underlying theme of "Yo-Yo" trading. Traders at the moment have a "sell first and ask questions later" mentality. Worries about the US economy continues to spook the markets. COVID-19 was an initial handbrake on economies around the world with the tech giants lapping it up or as @over9k would say "stay-at-home-tech" is where to be.

In conclusion
Even though the Hybrid Strategy tests well against the US markets "Yo-Yo trading still plays havoc with systematic trading" requiring nerves of steel.

Skate.

 

[ducati916]

Structurally the US market is different. The US has MM who will manipulate the stocks. What you will see is much higher volatility in individual names, which can shake you out of a trade based on a mechanical system. Going weekly avoids some of this but if the stock is selling off for real reasons, weekly can really multiply the loss. Earnings will/can also play havoc (that's probably true for Aus though).

jog on
duc

 

[Skate]

Structurally the US market is different. The US has MM who will manipulate the stocks. What you will see is much higher volatility in individual names, which can shake you out of a trade based on a mechanical system. Going weekly avoids some of this but if the stock is selling off for real reasons, weekly can really multiply the loss. Earnings will/can also play havoc (that's probably true for Aus though). duc

@ducati916 I'm under the opinion that trend trading strategies would work in all markets & your comments are a real eye-opener.

I had a few lengthy emails to explain how the strategy worked
The strategy tester found my coding & logic hard to follow at first. The report commented on how well the strategy performed in all markets tested, going on to say how well the strategy performed on inexpensive stocks in the US. The price filter is set from ($0.05 to $110) but my unique "positionscore" tends to select the lower price points when other filters & parameters come into play. The notation (performed well on inexpensive stocks in the US) at the time it wasn't important, as I had no intention in trading the US only the ASX market.

Skate.

 

[over9k]

Exactly - the volatility is just nuts at the moment. I did make a post a while back asking skate why he hasn't gone daily because of it.

There's quite a bit of money to be made just finding something choppy and throwing good-till-cancelled orders in

 

[Warr87]

Exactly - the volatility is just nuts at the moment. I did make a post a while back asking skate why he hasn't gone daily because of it.

There's quite a bit of money to be made just finding something choppy and throwing good-till-cancelled orders in

because he trades weekly and no system is every constantly in sync with the market at all times? cherry picking when to trade and not to trade, and cherry picking the timeframe you trade is not a systematic mechanical algo trader ...

 

[rnr]

@rnr, Thank you for following my paper trading and your concern. My account was in a drawdown of close to $3,700 when my system restarted producing signals after the March sell off. As I am not rebalancing my portfolio for this paper trading experiment and my position size is $1500 each position, the portfolio at this stage is fully invested at 18 positions. In fact I should have only taken 17 positions as I would have been short of funds for the 18th position.

@Saqeeb, thanks for your response and please be assured my comments regarding the development of your system are based on being constructive.

As your system was generating buy signals (without available funds) it became clear that you were not paper trading your system as it was designed, which is obviously up to you.

Some questions if I may:-

How long time-wise are you going to paper trade your revised system?

At the conclusion of this paper trading exercise are you going to:-

a) re-code your system, to reflect the method paper traded OR

b) effectively adjust the the results of the paper trading to reflect the system as initially coded?

The stats on the recovery from the resultant draw-down created by these two systems may be of considerable interest going forward.

Cheers,
Rob

 

[over9k]

because he trades weekly and no system is every constantly in sync with the market at all times? cherry picking when to trade and not to trade, and cherry picking the timeframe you trade is not a systematic mechanical algo trader ...

I never said it was all the time. I was implying that it might be worth it with volatility this high.

 

[Skate]

Exactly - the volatility is just nuts at the moment. I did make a post a while back asking skate why he hasn't gone daily because of it. There's quite a bit of money to be made just finding something choppy and throwing good-till-cancelled orders in

@over9k your logic to trade a daily strategy over a weekly system because of the recent volatility is well-grounded in theory but my research doesn't back it up. Trading a "Weekly Strategy" performs better with less work, less commission & less stress. (AFAIC)

My recent research found here
https://www.aussiestockforums.com/posts/1078832/

Weekly versus a Daily system
There are merits trading in both time frames but for "me" I prefer trading a weekly system, it also reduces the workload & stress. There is always a robust discussion when Daily versus Weekly trading is raised. I'm a believer if it works for you, it's right for you.

Posting in the "Dump it here" thread
I've always posted what I've found to be "the-best-for-me" when it comes to trading & if it's "best-for-me" it might also be the best for others but not always. Being consistent in my messaging I'm sure will help others.

It's worth repeating
If it works for you, it's right for you.

Skate.

 

[rnr]

it was theoretical, I use option 3, but this highlights the fact that this implementation requires a potential buffer of $ as you might (not to say often will and must) have winners still in your portfolio with unrealized gains

Hi qldfrog,

My apologies as I almost missed your response.
Yes, it is exceedingly difficult to re-invest unrealised profits. Perhaps that is why @Skate has an account called BANK in his system, thus making the task simple.
Hang on....just thought how you could get around the problem, yep it's a system account with the name AFTERPAY.

Cheers,
Rob

 

[Saqeeb]

@Saqeeb, thanks for your response and please be assured my comments regarding the development of your system are based on being constructive.

As your system was generating buy signals (without available funds) it became clear that you were not paper trading your system as it was designed, which is obviously up to you.

Some questions if I may:-

How long time-wise are you going to paper trade your revised system?

At the conclusion of this paper trading exercise are you going to:-

a) re-code your system, to reflect the method paper traded OR

b) effectively adjust the the results of the paper trading to reflect the system as initially coded?

The stats on the recovery from the resultant draw-down created by these two systems may be of considerable interest going forward.

Cheers,
Rob

@rnr, I have no concerns and I am sure your intention is for a constructive dialogue.

My system will keep generating signals every week, dependent ofcourse on the index filter. If I take the signal or not will be dependent on 1)if I am fully invested, in which case I cannot take any more postions, 2)if I have a sell signal, in which case I will have to replace the position with a new one.

So far I am happy with the system and I intend to paper trade for another 4-6 weeks to see if I will come across any issues. This will make it a 3-4 month time frame since the system's index filter has turned on. So far I have not had any issues with the system such as repainting or any other unforseen issues. To me, it is performing as intended and therefore I am comfortable to trade the system.

 

[Warr87]

@rnr, I have no concerns and I am sure your intention is for a constructive dialogue.

My system will keep generating signals every week, dependent ofcourse on the index filter. If I take the signal or not will be dependent on 1)if I am fully invested, in which case I cannot take any more postions, 2)if I have a sell signal, in which case I will have to replace the position with a new one.

So far I am happy with the system and I intend to paper trade for another 4-6 weeks to see if I will come across any issues. This will make it a 3-4 month time frame since the system's index filter has turned on. So far I have not had any issues with the system such as repainting or any other unforseen issues. To me, it is performing as intended and therefore I am comfortable to trade the system.

and try backtesting over the period you have been papertrading. the numbers should match up as well. I do this to give myself more credibility that the backtest results are at least close to what you expect on live data. Ideally you do 6months of forward testing like this but that is easier done if you are already trading a system.

my papertrading of the MAP system matched perfectly to the backtest results over the same period. even now I get the same affecting. peace of mind is important

 

[ducati916]

@ducati916 I'm under the opinion that trend trading strategies would work in all markets & your comments are a real eye-opener.

I had a few lengthy emails to explain how the strategy worked
The strategy tester found my coding & logic hard to follow at first. The report commented on how well the strategy performed in all markets tested, going on to say how well the strategy performed on inexpensive stocks in the US. The price filter is set from ($0.05 to $110) but my unique "positionscore" tends to select the lower price points when other filters & parameters come into play. The notation (performed well on inexpensive stocks in the US) at the time it wasn't important, as I had no intention in trading the US only the ASX market.

Skate.

Mr Skate, I think your system would work, overall, well. Some sectors, Biotech comes to mind, are however so news driven and the volatility so extreme (all year every year) that I doubt any mechanical system would fare particularly well. Whereas other sectors would really lend themselves well to a mechanical approach.

jog on
duc

 

[qldfrog]

I also believe you would need to refine your realm, either thru sector..as LeDuc noted, avoid Bio or even a filter removing stocks based on previous volatily/ size or even use ETFs only
But the sheer size and diversitg is tempting in the US

 

[qldfrog]

To clarify, when i use your realm, i imean the system builder, not specifically Mr Skate

 

[ducati916]

I also believe you would need to refine your realm, either thru sector..as LeDuc noted, avoid Bio or even a filter removing stocks based on previous volatily/ size or even use ETFs only
But the sheer size and diversitg is tempting in the US

The trend in much of the Tech. stuff currently would have loaned itself nicely to Mr Skate's system. From the bottom, its gone what, 100%+? I don't really know what the ASX has done, probably not that sort of return. Also where Mr Skate wants to place larger bets relative to ASX liquidity and has market moving issues, in the US that's like throwing a pebble in the ocean. You can have fills with zero slippage.

Re. ETFs, correct EMQQ as an example (which we hold) has the returns, without the single stock risk. XLK another index (Tech) that provides the returns (currently) without single stock risk. You also avoid issues during earnings. There are probably as many ETFs as there are listed ASX stocks. You'll never run short of candidates. Also, lots of inverse ETFs which allow a short strategy without the risks of actually shorting the market.

Also far easier to hedge: either via inverse ETFs or Options.

jog on
duc