qldfrog weekly Skate inspired system

[qldfrog]

moving into daily systems causes a few starting pains:
I use overseas indexes and just realise that when the US market closes, it is still too early to download Premium Data, Mexico and a few others are still live for one hour;
Premium Data world indices are only ready after 7AM Qld Time.That explains some "repainting" I saw yesterday..
probably explains what I saw as code bug last week:
no repainting, just not fully updated data.
If this can help more beginners

 

[qldfrog]

more elegant than my code indeed if this works to your preferences

let's have fun, based on @Sir Burr improvements, an all in one version:
//get proper Asx Tick Price
//rounding up if up is true, rounded down otherwise
function AsxTick( price, up )
{
factor =
IIf( price < 0.1, 1000,
IIf( price < 2, 200, 100 ) );

return IIf( up,ceil( Price * factor ) / factor ,
floor( Price * factor ) / factor );
}

 

[qldfrog]

End of the week, some positive, some less so;
My daily system started on monday:50k now nearly fully invested:
by the end of week: profit of $1542 or 3.09% which is basically XAOA

QFDuc not so lucky and lost 1k just on today :
and turned red: not good:

QFDuc started 1/05/2020
Total invested $ 90,009.85
Cash remaining $ 15,619.62
Current portfolio value $ 89,799.72
Profit -$ 210.13
Nb of active positions 17
Nb max position 20
Return since start -0.23%
Annual return so far -4.06%
purchase value per position $4,489.99
Invested percentage 83%
On monday, 8 out of 17 positions are for sale and 11 more will be purchased, tell me about dynamic...
--ZL extra:
better $770 profit for the week but is beaton flat by the XAOA
yet it is only 35% invested and it will take time to ramp up, I am aware of this

ZLExtra started 1/05/2020
Total invested $90,000.00
Cash remaining $58,874.07
Current portfolio value $91,177.20
Profit $ 1,177.19
Nb of active positions 7
Nb max position 20
Return since start 1.31%
Annual return so far 22.73%
purchase value per position $4,558.86
Invested percentage 35%

On monday, 3 buys only
overall? impressed by my daily beginner, not so by what should be the bread and butter
Luckily, Mr Skate saved the week with a very nice 1.1k profit for the week overall 711$ in 2 weeks from 20k investment, very promising.
Summary for my system plays this week:
1.1-0.6+0.77+1.5 or around $2.7k ....or above 1% return...not bad

 

[qldfrog]

Sorry for the delay, busy time with 2 daily systems initiated, Skate following and my 2 weekly still getting incremental improvements while life..



yesterday at open ZL extra purchased :

FAR 281250 @ $0.0200

JLG 1786 @$2.5000

TPM 559 @ $8.0400



QFDuc sold:

ANN 133@$34.63

APX 145@$30.40

AST 2334@$1.81

BKY 14062@$0.26

CDV 10344@$0.42

DTL 969@$4.52

HTA 29032@$0.15

RMS 2603@$1.71

RRL 800@$5.27

for a loss of $1496, the dynamic side does not seem to pay in the current market

and purchased :



ALK 5113@$0.87

AQZ 1577@$2.79

BRN 73333@$0.06

CGC 1419@$3.22

JLG 1800@$2.49

OGC 1371@$3.28

PNR 23684@$0.18

PNV1685 @$2.67

PRU 3543@$1.26

RBL3474@$1.27



to date so midweek today, QFDuc is 94% invested:

should be 100% but I did order twice the same packet by error and add to sell back one of the 2, for a small profit , i will catch up next week but Back tests will differ a lot

and we have a 2.7pc profit since 4/05 in what is a booming market so QFDuc not that great

ZL extra similar return but 50% invested so better risk return ratio by far, nearly double

Daily system going well: up 8% from a 18/05 start, fully invested let's see how all this firms up when the inevitable pull back arrives

 

[qldfrog]

another great wek:
Daily:
second daily system engaged 2 days ago, fully invested:+1.7%($330);
main daily system 85% invested, +4.67%(+$2.4k)
Skate system saw one parcel sold by error, good week:+5.5% (+$1.4)

Weekly QFDuc:19 out of 20 positions , +3.12% or +2.8k this week;
trailing the xaoa still:
QFDuc started 1/05/2020
Total invested $ 90,009.85
Cash remaining $ 5,223.26
Current portfolio value $ 92,607.92
Profit $ 2,598.07
Nb of active positions 19
Nb max positions 20
Return since start 2.89%
Annual return so far 37.63%



Lastly:
ZL extra: half invested+0.8% or $700
not that great well behind the xaoa but this is a strategy which is not supposed to parallel the other so fair;
low risk 9we hope so,and more stable
Time will tell

ZLExtra started 1/05/2020
Total invested $90,000.00
Cash remaining $45,384.71
Current portfolio value $91,886.21
Profit $ 1,886.21
Nb of active positions 10
Nb max positions 20
Return since start 2.10%
Annual return so far 27.32%

 

[qldfrog]

nice experience installing my system on aws;
Get an aws account, create an EC2 window server instance and run it;
then install my data provider (Premium data) but you might need to allow IE download enabled as it is off by default;
so download installers, install then on AB site, get your AB installed and the plug in setup;
When this is done, you are basically cruising, was much easier actually than i thought,
next stage: automatically update data, then run my various explore in a batch, sending the results in an email for action

 

[InsvestoBoy]

nice experience installing my system on aws;
Get an aws account, create an EC2 window server instance and run it;
then install my data provider (Premium data) but you might need to allow IE download enabled as it is off by default;
so download installers, install then on AB site, get your AB installed and the plug in setup;
When this is done, you are basically cruising, was much easier actually than i thought,
next stage: automatically update data, then run my various explore in a batch, sending the results in an email for action

Why pay for an EC2 for just a single VM?

Expensive bandwidth, core pricing, etc.

Check out AWS LightSail, can get a much cheaper single VM with much cheaper bandwidth.

 

[qldfrog]

All what i need is free for a year

 

[qldfrog]

got a workable setup for an overnight stay at the beach tomorrow:
whole aws server running, updating Premium data running AB and my systems all good.
Nice to have especially when borders reopen and Caribbean beaches are a destination again , for the time being staying in Qld..
.


quite happy...
need to add a bit of automatic update and explore results exports


Back to the hard yaka:
had a look at the overall performance of systems over the past week, including exposed capital etc, I lost against the XAO which is not that good
It is always worthwhile looking at the big picture and not focusing on a few numbers;
Since inception:
daily eft beat index..(just started last week so not really meaningful)
daily generic beating index 7.9% to 5.7%..(started 18/05)
weekly qfduc beaten flat by index 10.35% to 2.9%..(started 01/05)
weekly ZLextra beaten flat by index 10.35% to 2.1%..(started 01/05) but with low participation(<50% invested) so far so overall better performance than QFDuc

ZLextra has seen major code changes since inception which will take effect from tomorrow to increase participation and nearly double returns with low extra DD based on backtests
Time will tell

 

[Warr87]

nice setup! the AWS's look a bit expensive depending on the context. Are you just paying for 'on demand'? I've never used them so not entirely sure how it works. Good setup though.

As for your systems. Is it possible your systems are 'out of sync' with the market? Not surprising as it is chaotic. A mixture of bulls and bears, and enough speculation to confuse anyone. Things certainly don't seem 'normal' right now. Over the long run I'm sure you will outperform the index. (I certainly understand your annoyance though.)

 

[qldfrog]

All the above aws aws is fully within the free tier limits, it costs 0, fully free on the first year and i expect it to remain so based on past experiences with other test platforms..not share trading..i did before

 

[qldfrog]

System wise, i tried to implement 2 systems with different behaviours.
Not great results for the weekly but nothing to be ashamed of especially after zlextra work in the last few weeks which should improve overall results

 

[qldfrog]

A great week and I should have my fuses ready for my weekly system on Monday.
I changed my index comparison to XNT as XAOA is sometimes slow to be released on friday night
the week that was..
Daily #1:+6.27% overall this week, overtaking the XNT since inception 15/05;
Daily #2: +7.62% overall this week, overtaking the XNT since inception 22/05;

SKate system: I made another mistake when one parcel was sold as it was sharing the brokerage account of another of my system:
overall lost 6.7% this week (Gold miners?), but could be due to my own management mistakes as i am now only 19 positions invested;
QFDuc +1.41% , still below XNT performances
ZLExtra +2% , still below XNT performances
overall all systems:+2.4% as I am heavily overweight weekly system (nearly 1 to 2)

 

[qldfrog]

adding fuses to weekly systems;
With ever increasing speed in the market changes, I was feeling uneasy with the concept of being fully passive during the week whatever happened and only react on the next Monday open;
the principle:
at the end of the week as I determine buy sell, I also update SL which can be activated anytime automatically during the week based on various criteria;
If any of these is activated during the week so be it and we will buy again on the next Monday.
While not that easy to backtest , and acknowledging some of these SL sales will fail, it seems to drastically increase the results of at least one of the system, and create minor improvements for the second.
There is also definitively a Peace of mind advantage.
Will prepare all that this weekend and enter the conditional orders for the next week

 

[qldfrog]

SKate system: I made another mistake when one parcel was sold as it was sharing the brokerage account of another of my system:
overall lost 6.7% this week (Gold miners?), but could be due to my own management mistakes as i am now only 19 positions invested;

Errare humanum est : Mr @Skate , the loss this week was just -2.49% so not really worse than the system, I had the wrong SP for SLR
My apologies for the previous mistake, the Action system is going reasonably well

 

[Warr87]

A great week and I should have my fuses ready for my weekly system on Monday.
I changed my index comparison to XNT as XAOA is sometimes slow to be released on friday night
the week that was..
Daily #1:+6.27% overall this week, overtaking the XNT since inception 15/05;
Daily #2: +7.62% overall this week, overtaking the XNT since inception 22/05;

SKate system: I made another mistake when one parcel was sold as it was sharing the brokerage account of another of my system:
overall lost 6.7% this week (Gold miners?), but could be due to my own management mistakes as i am now only 19 positions invested;
QFDuc +1.41% , still below XNT performances
ZLExtra +2% , still below XNT performances
overall all systems:+2.4% as I am heavily overweight weekly system (nearly 1 to 2)

I saw my system hit new highs by wednesday and then evaporate on Thursday (losing last weeks profits too). Will do a proper update tomorrow but I think a few people lost some open profits this week ....

 

[Trav.]

With ever increasing speed in the market changes, I was feeling uneasy with the concept of being fully passive during the week whatever happened and only react on the next Monday open;

@qldfrog I have been thinking about your post and possible options but have been unable to come up with anything to help with the feeling of unease whilst trading a weekly system.

Then I thought why are you even in the market when you have these feelings. So may I suggest

1 - Leave the code as it is and stop tinkering
2 - Park the weekly system for a a bit and concentrate on the daily systems that you

- Maybe test / use closing out all you positions on Thursday and re buy on Monday if a still valid buy.​

Anyway just my thoughts which I hope you don't mind me posting.

Cheers

Trav

 

[qldfrog]

@qldfrog I have been thinking about your post and possible options but have been unable to come up with anything to help with the feeling of unease whilst trading a weekly system.

Then I thought why are you even in the market when you have these feelings. So may I suggest

1 - Leave the code as it is and stop tinkering
2 - Park the weekly system for a a bit and concentrate on the daily systems that you

- Maybe test / use closing out all you positions on Thursday and re buy on Monday if a still valid buy.​

Anyway just my thoughts which I hope you don't mind me posting.

Cheers

Trav

@trav, all feedback welcome
The solution I mentioned solves my problem:
a combination of intraday stop loss , with trigger based on weekly read index/indicator triggers
Similar towhat I do in a normal weekly system but activated intra day;
To be even more precise, the SL trigger only changes weekly so no major extra work: just set/adjust SL once a week on portfolio packets

If you remember, I tried to do SL just before the crash then removed them; I so went from +20k or so to -20k just after the february fall.With a current GTFO , I would have lessen the losses, with this in place , I would have left a winner
If it helps:
2 backtests showing the smoothing and better performance:
QFDuc without the intraday SL:

with the intraday SL

as Mr Skate says, BT are bull****, but when the change is actually allowing you to sleep better, why even bother trying without, there is slippage factored in the SL and the above graphs are the more dramatic but ZLextra benefit as well
Happy with the daily systems which are going gangbusters and managed to detect IFN, BRN, AYS etc
Only problem is that I get left behind at the open for some buy, but can not enter my orders high enough in the previous evening
Have all a great weekend

 

[qldfrog]

just need to add: I have plenty of time right now as pending sale of my PPOR, I am in stand buy and unsure if I want to restart my company and how/in which field while international travel is a no go.
I ended losing a total of $18k with my systems by end of march. a learning fee
To date, the new or updated ones are up 18.2k, I broke even this week systems wise and I will try to preserve this gain.

 

[Trav.]

All good mate as long as you are happy with where you are at.

On a side note I read this today which is relevant to me and could be thought provoking for you as well.

Quantitative Trading Systems - Howard B. Bandy