qldfrog weekly Skate inspired system

[willy1111]

Hi willy1111,I traded for a sort period in my super fund and was profitable. The only down side was the lack of stocks to choose from and the max % allocation allowed to those stocks. In the end i gave up as many of my stock picks weren't allowed.

Also i run my trading as a business so capital gain isn't an issue.

Cheers.

I do mine through an smsf administered by esuperfund for approx 900 a year plus ato levy approx 250 plus annual fee for company of 50 so about 1200 a year regardless of size of fund...this allows me to use Commsec the same as any other Commsec user so no restrictions on stocks etc.

Admin is super easy too.

Only issue for me is I'm a long way from preservation age

 

[willoneau]

That is interesting willy1111, i did it through AustralianSuper at the time.

I would be interested to know what your yearly costs are before any trading takes place. I know commission is a % above any lot above $10000.

 

[willy1111]

That is interesting willy1111, i did it through AustralianSuper at the time.

I would be interested to know what your yearly costs are before any trading takes place. I know commission is a % above any lot above $10000.

Yes quite a few superfunds have a direct investment option where they allow one to pick their own stocks generally within ASX300 index.

I tried that and found it too clunky and limiting hence went the smsf route.

Mine costs about $1,200 per year before trading costs, no matter if you have a $10k balance or $10m balance. Costs vary depending on who administers it for you, an online provider like esuperfund who I use is about $1,200.

Commsec is normal commission.

 

[willoneau]

Thanx willy1111 i might look into it.
what about the costs of a SMSF

 

[qldfrog]

very bad week:-1.1 k including a 1k loss just on today....
Total invested $ 100,000.00
Cash remaining $54,391.89
Current portfolio value $105,550.53
Profit $5,550.53
Nb of active positions 9
Nb max position 20
Return since 25/02 5.55%
Annual return so far 13.42%
purchase value per position 4975.368182
Invested percentage 48%

some action on Monday
Have all a great week end

Report Bookmark

 

[jjbinks]

hi jjbinks, as skate said in a post - back testing means jack, I agree totally and only use it to determine if I have positive expectancy and not to really look at returns. A robust strategy should be able to withstand the ebb and flow of the market over the long term, Nick Radge states many good systems can take up to 3 years to become 100% profitable depending on when they are started in the market cycle.

Really what I was getting at when I asked about backtest results was that how it fits in with monitoring of your live trading results. As a trader you must have some kind of evaluating your own performance and my question is what do you use as your benchmark. As you mentioned different systems can take more or less time depending on market conditions to become profitable. (Not sure what you mean by 100% profitable as opposed to say 50% profitable). If you back test over periods which cover different parts of the market cycle this should give you an idea. I feel 3 years is a long time to wait to find out if your system truly works or not!

I guess, ultimately that leaves you with trying to find a system which becomes profitable within an acceptable time frame for you. For me that would be somewhere between 6-12 months! But it is easier said than done.

 

[qldfrog]

I fully disagree with the notion backtests are crap
As discussed, backtests are equaling my real system since i started, you backtest different phases..up down,going no where shift dates by a few days and get a pretty good idea of your system value.
In my opinion, if you have to wait months to see if a system works, then you do not program it properly on backtest or you are in a new never experienced before market
So a lottery
Backtests do not represent an image of the future we all agree, but they are critical in a pure systematic system without human based decision

 

[peter2]

@qldfrog I understand your disappointment with your portfolio performance so far. I would like to remind you of something. Your portfolio is only 50% invested, so you should only expect to have 50% of whatever the benchmark has done. I see the XAO is +12%, 50% of this is 6% and your portfolio sits at 5.5%. Not bad at all.

We could discuss your reluctance to invest more as the market rally continues.
We could discuss your entry criteria which I thought was a little late (like all momentum strategies).

But as you're monitoring your back testing and your performance is matching nicely, I think you're going along quite well. Your conservative nature will be rewarded when the market dips as it always does.

 

[willoneau]

Hi jjbinks, let me explain first what i meant by 100%, it was in reference to going broke, just not really explained well. Nick Radge explains it better in Unholey Grails.
I didn't say back testing was crap i said as skate pointed out means jack, meaning past tests will not be the same as future results. Doing Monti carlo simulations points that out.
The biggest and most important aspect of back testing for me is does it have a positive expectancy over time.

 

[willoneau]

At this stage jjbinks my benchmark is to be profitable at the end of the financial year. I know i want to be able to monitor my progress and will be looking at doing that in the near future, just not sure how to track closed positions, live positions or both together.

 

[qldfrog]

@peter2 thanks, yes i am aware and i like being even with xao with half the risk, just a bit frustrating to realise all this initial work of system setup and research could have been match by VAS
But i am old enough to know that the market, or the world actually, does not care with work effort or justice etc
Real frustration is on having only 24h in a day and not having time to work and design a more dynamic second system to take advantage of what has been a stellar unusual market since January

 

[willoneau]

@peter2 thanks, yes i am aware and i like being even with xao with half the risk, just a bit frustrating to realise all this initial work of system setup and research could have been match by VAS
But i am old enough to know that the market, or the world actually, does not care with work effort or justice etc
Real frustration is on having only 24h in a day and not having time to work and design a more dynamic second system to take advantage of what has been a stellar unusual market since January

Hi qldfrog, I did some back testing over the last couple of years and found my daily system out performed my weekly.But over all my weekly was way ahead, my thinking was to incorporate the two for times in the market when these changes happen and also allows me to trade more often which indulges my passion.

 

[jjbinks]

Hi jjbinks, let me explain first what i meant by 100%, it was in reference to going broke, just not really explained well. Nick Radge explains it better in Unholey Grails.
I didn't say back testing was crap i said as skate pointed out means jack, meaning past tests will not be the same as future results. Doing Monti carlo simulations points that out.
The biggest and most important aspect of back testing for me is does it have a positive expectancy over time.

I see what you mean re the 100% now. Completely agree that looking at back test without monte carlo results is essentially useful.
I think we don't talk about monte carlo results in our threads as much as we should but in someways for novices like me it perhaps it is one of the most useful results to guide expectations. Definitely something I paid attention to before starting trading. But have not been thinking about MC simulation much recently so thanks for the reminder!
Obviously expecting to match results from a single back test is stupid. But if you run your test over different time frames MC give you an idea what your chances of profitability is over different time courses.

 

[willoneau]

The MC simulations use the same time frame just alters the criteria slightly ( you can adjust how much deviation in amibroker, I think i used 2%). Run 1000 simulations and see best and worse results. If still positive then on a winner.

 

[jjbinks]

The MC simulations use the same time frame just alters the criteria slightly ( you can adjust how much deviation in amibroker, I think i used 2%). Run 1000 simulations and see best and worse results. If still positive then on a winner.

You can run MC over different time frames and durations.
e.g. 6 months, 12months, 2 years etc and see how long it takes for you to be profitable 95 or 99% of time on MC.
also you can chose different periods e.g. test during GFC and other times to see how your system performs. I think this then gives you a bench mark to assess if your system is working real time and edge is valid.
e.g. MC over multiple different 12 month period predicts 5% or greater return 95% of the time and your return is less than this I would say that you have a problem.
In my experience getting really tight MC results over 6-12 month period is very hard. 2-3 years for most systems they should be profitable. (This goes back to what you mentioned about radge saying that all good systems should be profitable within 2-3 years)

 

[willoneau]

That sounds like alot of work and the market is always changing so not sure of the benefit. For me risk of ruin less than 1% within 3 years with positive expectancy is good enough to start forward testing a system.

 

[qldfrog]

I just fulfilled my first promise: comparison of income tax rates among a few selected places, demonstrating if need be that we are horribly taxed here yes are supposed to go to private hospitals, pay for education and offer poor expensive public transport
https://www.aussiestockforums.com/threads/income-tax-rates-in-the-world-vs-australia.34841/
Now need to do a thread on becoming retiree and work on system number II
Have all a great week

 

[willoneau]

I just fulfilled my first promise: comparison of income tax rates among a few selected places, demonstrating if need be that we are horribly taxed here yes are supposed to go to private hospitals, pay for education and offer poor expensive public transport
https://www.aussiestockforums.com/threads/income-tax-rates-in-the-world-vs-australia.34841/
Now need to do a thread on becoming retiree and work on system number II
Have all a great week

You thinking of leaving Aus qldfrog?
does the smaller size of Aus population have an influence on tax% compared to other countries with larger populations?

 

[qldfrog]

You thinking of leaving Aus qldfrog?
does the smaller size of Aus population have an influence on tax% compared to other countries with larger populations?

No, pure mismanagement
Yes thinking of not being a tax resident
But i expect very low income now that we retired so oz tax might be bearable
Just have to cry for the country ..

 

[qldfrog]

o with the system, i purchased today 3148 JLG @1.565 for a total of 4937$ inc brokerage
I now have 10 positions out of 20 after 6 months of live play