This is a mobile optimized page that loads fast, if you want to load the real page, click this text.

qldfrog weekly Skate inspired system

Jump to new
no dilemma: flying bat stay roosting in its tree: no buy action today
 
qldfrog said:
do i stay paper or start live?...my dilemma
Click to expand...

@qldfrog, paper trade would be the way I would go until an edge has been proven with out of sample data. Posting the signals before or after the market opens makes no difference but displaying the signals would allow me to follow more closely.

Skate.
 
it is actually a bit scary to start something not based on a known indicator, i am no genius so it is probably a mistake..
 
Skate said:
@qldfrog, paper trade would be the way I would go until an edge has been proven with out of sample data.
Click to expand...
qldfrog said:
it is actually a bit scary to start something not based on a known indicator, i am no genius so it is probably a mistake..
Click to expand...

Dot the "i's" & cross the "t's"
Attention to the details when coding a new strategy is of the utmost importance. Mistakes in this game cost money. The trait of having a go needs to be acknowledged but I've noticed confidence & positivity are not your strongest suits.

Skate.
 
Positivity definitively not, confidence hum :not absence of, just realism.
 
OK so week 4 or so of the disaster series:
another month like that and even TD will be a better option for this FY :-(

So daily
DLQFDuc closed on the 19/05 with a gain of $1.3k or 1.3%..I am closing my winner
DLGuppy: new version implemented and now fully invested with $3.1k loss or 5.8%
[a full $1.8k loss due to ARV..hate that as my sell order was canned by the trading halt and then crashed..hum, BTP ,AJM, and now ARV..I seem to like these multi loosers]
volatility:
out just before they jump and in before they crash..by 1 day...
volatility US stable
volatility ASX -$0.6k or -1.6%

weekly:10% invested only
-$0.6k or -0.57%
Overall in red:
-3k or around -1%
 
a diversion this morning trigerred by Mr @peter2 posts

Ranking by alphabetical order?
That looks crazy but i noticed that on my daily systems,i often go into a kind of cycle:
Today is buy of the M day, then tomorrow buy of the P days
Looking like the market conditions triggering my purchase follow a kind of alphabetical order.
And no i do not limit the explore ourput numbers,just ranking
Can be a feeling, or we could have a real factor here.
Just throwing ideas: analysts could for example look at a set of investment candidates and split their work for the week in 5 packets based on alphabetical order.
That could make sense,then after they buy their shares, my trend systems would pick this up and flag these codes as buy?
Should I stop the Shiraz?
 
qldfrog said:
Should I stop the Shiraz?
Click to expand...

YES - stop drinking

Amibroker Ranking is called - PositionScore
@peter2 acknowledges there are always more candidates than the available money & most times more signals than required to fill a portfolio. From my research alphabet ranking doesn't produce "bang-for-buck" neither does Radge's "Bang-for-buck" PositionScore.

A Gem from Nick Radge's Bang-For-Buck article
"You will therefore find that capital usage in higher-priced stocks is inefficient compared with lower-priced stocks" Nick is suggesting that efficiencies can be garnered by taking the "lowest-priced stock" first.

With PositionScore it's horses for courses
It's about dancing to the music being played, meaning one PositionScore doesn't fit all strategies (they need to be in tune with the strategy). The PositionScore can be one of the deciding factors of the overall profitability of any strategy. At all costs, you need to get the ranking correct or you might just throw away a good strategy or trading idea.

Skate said:
The Simple Strategy "PositionScore" is entirely dependent on the "ROC".
Click to expand...

PositionScore
The PositionScore decides which trades should be entered if there are more entry signals than the maximum allowable number of open positions or available funds by ranking them in numerical order. In such a case, AmiBroker will use the absolute value of PositionScore to decide which trades are preferred.

peter2 said:
How can I reduce the number of opportunities to a manageable level?
(i) Prioritise the stocks with the lowest price (like @Skate)
Click to expand...

For the record
I use a variety of ranking methods as the ranking (positionscore) needs to be in tune with the objective. Using the lowest price first has its advantages when dealing with defined indexes. Trading lower price positions can springboard the returns due to the power of percentages.

Backtesting alleviates the need to guess
"PositionScore" affects only the backtest & optimization, this is the reason we need to add a filter, without an additional filter it wouldn't display the ranking in the Exploration Analysis. The "PositionScore" is important when it comes to the profitability of any strategy.

Skate.
 
Off for a coffee break but will reply tonight i was actually refining position score this morning.. coincidence.
It is an area i do not feel i am confident having found the grail yet
 
ranking:
the obvious one is of the form 100-C so giving a preference to lower penny stocks, and you benefit from the huge in % increments..but it usually adds volatility;
another issue is that on generous selection you end up with dozen of 4c stocks or similar; so kind of random selection?
Lastly you basically cycle with a very small set of stocks, the smallest ones
yes it gives good backtest but not that happy with it;

ROC as Skate noted is worth a look;I used it in conjunction

you can try Nick Radge or variation:
PositionScore =((10000/Close)*ATR(period)); decent but never the best in all my trials

and then you can use the specific of your system:
if you wait for A>B to buy, is the Abs(A-B ) value meaningful and dictating preferences?
and all variation of these including TSI indicator values etc etc
Sadly I have never yet found a position score i am happy with.
This is probably my missing link
 
I agree with skate on the issue of forward testing with paper or miniscule amounts of money until you're absolutely certain the system has a consistent edge.

Have you seen the flying bats that get caught in the overhead power lines? Not a pretty sight.

 
thanks Peter, there is a reason I did not name my system the soaring eagle...
 
Good evening all, and May is coming to an end;
Good end of week today for the frog:
daily:
Guppy: +1.2k or +2.34%;
volatility ASX: +2.3k or +5.2%
volatility US: +0.5k or +2.9%
weekly qfduc: -0.6k or -0.6%
disappointing as after a great start till december or so beating or matching XNT, just eroding gains weeks after weeks...
Will need to check if any change has been costly or system unsuitable for choppy waters
and the joker:
starting Flying Bat with slow entries (restricting entries to 4 a day):
now nearly 90% invested and at a $1.7k loss or -2.08%

overall:1.7k profit or +0.6% 80% invested by today.
Have all a great week end
 
Good evening all, and a relatively great week until today where my small caps got crashed..
so lost $1750 in a day where the asx was going higher


daily:
Guppy: +$3.9k or +7.5%;
Flying bar in its leaving the tree mode:
-$3k or -3.8%
volatility ASX: +$1.7k or +3.6%
volatility US: -$0.15k or -0.9%
weekly
qfduc: +$0.6k or 0.6%

overall: $3.05k profit or +1.02% 95% invested
give me 52 weeks like that
Have all a great week end






Quote Reply

Report
 
qldfrog said:
you can try Nick Radge or variation:
PositionScore =((10000/Close)*ATR(period)); decent but never the best in all my trials
Click to expand...
Maybe try this against random too...

Turnover = C*V;
Rank = ROC( Turnover, 1 );
PositionScore = Ref( 100 + Rank, -1 );
//PositionScore = Random();
 
Good evening all, and a really bad week


daily:
Guppy: -$3.3k or -5.9%; disappointing
Flying bat
-$2.6k or -3.5% but we start having some good winners on older purchases
volatility ASX: +$0.25k or +0.5%
volatility US: +$0.2k or +1.2%
weekly
qfduc: -$0.4k or -0.4%

overall: -5.8k loss or -2% 94% invested with a lot of movement this week
I wonder the impact of trying to reduce exposure to specific stock:
If an explore asks me to purchase a specific stock for a system, but that stock is already present in system 2 or 3, I bypass it and purchase the next one.
in a so so market, it probably means I take far too many not optimal stocks.
as i have 3 systems, most days, I bypass quite a few then get sub optimal choices that my dailies discard on the next day with a 4 or 5% loss
anyway, luckily, renovations are ending and I should be able to spend a bit more time on the systems
 
Note.my current systems are heavily invested in small caps..aka in australia junior miners and exploration companies which is not sustainable long term as I delay my planned new systems due to time constraints
so I need to get cracking to add stability by numbers
 
Today see the start of my last systems:
I have two volatility daily systems (US/ASX) stable and unchanged
2 other dailies
DLGuppy: relatively robust proven methodologies, in red with quite high volatility
The current version is now quite mature but even in backtests is in draw down since April.let's be patient
DLFB: Flying bat very green and own mix, will be loosing money in the first few weeks, hopefully should stabilise this week is the market is clement. Current version is in red both backtest/real.

weekly:
QFDuc: very disappointing lately:lost 3% today..yeap, and is in 10% DD from end January

lastly just started QFSec slightly influenced by sectors..as much as i can do that here on the ASX ; and otherwise a classic trend following system using Momentum, AAron, TSI
Hopefully will decouple a bit from my other systems;
Last week I lost heaps due to my avoiding more than one packet of the same stock among strategies: I removed this..at the very least , I can now check vs backtests.
.but it has a risk management consequence: more exposure to single stock crash, etc
short night the day the darling of the market collapses...
This is also a good time in the fiscal year for me to start the systems as the ramp up is expected to be in the red, and i can offset these loses..but not so sure it is good timing in term of markets...
I also add today some thought about long term investments: see my ammonia thread
 
OK, the end of the bloodbath week at last: The red wedding......
My systems generally target lower priced stocks, with higher volatility..that is the way you can beat an index, so fair;
In Australia, these are small caps : exploration shares or junior minors overwhelmingly; and when resources get smashed, these get double or triple smashed, including the safe haven gold and PM.
So an horrendous week and a VERY BAD timing to start my last weekly on tuesday....
Thankfully, i decided to only enter 10 stocks and not 20 as a timing mitigation risk..that probably saved me an extra 2%

The carnage got mitigated by a $3k rebound today for the few shares remaining hold :
a good argument to mix daily and weekly system: daily to react fast in big crash (this was not a crash IMHO), and weekly to smooth days like the ones we saw
The dailies:
Flying bat:-5.8k or -8%; 93% cash
Guppy -1.3k or -2.4%; 88% cash

volatility ASX +0.9k or 1.92% now fully cash

volatility US $-155 or -0.9% now fully cash

Weekly:
the new kid on the block:
QFSec -1.8k or -2.3% still trying to improve its Sector leverage, obviously this week was just a sea of red
QFDuc -0.4k or -0.4%

overall -$8.6k or -2.4%
It does hurt (a lot) but I dare say it is not especially bad vs similarly exposed stocks
Poor Mr ATO, I might barely break even this FY with my shares between bonds, USD and investments/systems..will do a full report in july
Enjoy the week end
 
You must log in or register to reply here.

Similar threads

Trading systems/methods guidance
Replies
14
Views
6K
rnr
R
Emergency Exit for Weekly System
2
Replies
27
Views
5K
MovingAverage
Swing System
Replies
19
Views
5K
MovingAverage
Weekly Breakout Strategy For QQQ ETF
Replies
17
Views
5K
KevinBB
K
System optimisation: IS training vs. OOS testing, and Paper Trading
Replies
10
Views
2K
Garpal Gumnut
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more...