Australian (ASX) Stock Market Forum

Weekly Portfolio - ASX

Jump to new
Something wasn't adding up with my profit when I logged into my account, and my systems return. I think my spreadsheet is taking into account my original funding for the account, not what it is now. i thought I had this sorted. Needless to say my returns are well inflated. I will get around to fixing it, but just half it and that is what I am at now. Still a good return, and still well above the XAO.

Sorry for the confusing and unintentional misleading. I knew I wasn't that good ;).
 
Despite finally being on holidays from work I have found myself pretty busy. Wont be posting an update just yet as I haven't got around to fixing the performance tracking issue just yet. It's on my to-do list.

The EA's are doing well. I even started the EA's trading CORN and SOY futures on my live capital account (been in my demo account for at least a month). They are also trending very nicely right now too.

An update for the SuperASX300 will also be coming. Some time this week.
 
Week 48 (late update)

Buys: ERM
Sells: SDG (stale)

SDG came up as a stale exit. this didn't actually sell until today. I had to keep checking it as there was just no volume on this thing. It was a little frustrating. It did end up closing for a +68% profit, so can't go wrong there even though I had to keep lowering my sell price.

It was a good week last week. The momentum in my profile is still growing. I'm gaining pace greater than the market which is always nice. I finally got around to fixing (in part, I'm really not sure how accurate it is) the performance tracking. I will be more careful in the future. It got to the point where the performance stats just weren't matching up to what I expected and weren't making sense. It should have been obvious earlier ...

Revised performance charts.

at +31.5%. Still very healthy and something I am very happy with!

1610000465202.png

1610000493678.png


As for the CFD trades. That is going great. As of today it is +21.5% realised with another +22.7% in open profits. My donchian bot is doing good. I also started gold spot price, thankfully before its up-move earlier in the week. Though it did get stopped out. The ETH bot has been stoped out from a few false breakouts but is currently doing well.

What I'm now trading with CFD's using an EA:
XAUUSD (spot gold)
BTCUSD
ETHUSD
CORNF
SOYF
AUDUSD (this is the only one not using the donchian EA)

The Super account will get an update of its own soon. Only entered into the trades yesterday by the time settlement came around and my money was available.
 

Attachments

  • 1610000481039.png
    1610000481039.png
    26.6 KB · Views: 1
Week 49

Buys: none
Sells: none

I'm now at +33.4%. I think others had some better results this week. I'm still happy. No real comments on the system or the week. It pretty benign.

1610098754758.png
1610098766663.png

1610098813615.png

1610098828201.png


Other updates:

Happycat is +11%.

CFD
+26.5% with +15.5% is open profits still on the table. Markets have been trending in the ones I now trade (SOY, CORN, Spot Gold, BTC, ETH, and AUDUSD). I'm pretty happy with how its progressing. I may add more capital to this soon as it was seeded with an initial $1000.
 
thanks mate. i'm still suspicious of the results, but I think I'm just second guessing my success thus far. The results weren't as stand-out until I compared them to the XAO.

I just ran a quick backtest over the same period. I'm lagging significantly behind that backtest (by half). But I missed a few trades and struggled to get a few fills when I sold.

This should also be one of my worst years, I can only hope. With the market crash last year I'm hoping for strong market returns this year. Time will tell :).
 
A little bit of psychology, a little bit of commonsense and a huge pat on the back. A trading system will only earn what it's designed to earn. Our role is to apply the system as perfectly as possible in order to earn the maximum that's available. If the system is not fully automated then we're going to make some mistakes. Overall, mistakes cost us and reduce system profit. Hopefully the system has enough of an edge that a few mistakes won't cost us too dearly. I don't like that you're only earning 1/2 of what your system should earn. That's a huge problem that you'll have to fix. The system could be at fault if the buys and sells are frequently impossible to get in the real market . Or you're the weak link in the system. You've got to minimise the number of mistakes.

I think you're correct in comparing your results against a system back test (or a forward test). Don't fall into the trap of comparing the results of your system against results of other systems. If you see that other systems return better results then work on developing better systems. @Skate is a master at this.

Don't overlook the significance of your trading achievements. Your MAP system started at exactly the worst possible time. Did you quit? No. You traded your system through a very difficult period and now you're starting to see the rewards as your equity curve leaves the XAO in it's dust. Very few new traders experience the immense personal satisfaction that you should be feeling now. Most of them quit or change systems. You have dug yourself out of a significant drawdown. Having done it once, you must know that you can do it again.
 
About the lag against backtest: i believe check against backtest is indeed the criteria to track.i stopped one of my system due to unexplained delta.
One way i have explained deltas which can seriously affect performance is when i have both buy/sell at open but no cash reserve left.
my broker does not accept buy order if i do not have cash or pending cash available, so i need to wait for some packets being sold before entering an extra buy order.
Based on sell execution orders at open, it sometimes mean my buy orders are late and miss the open price.
But 50pc underperformance is huge.worth investigating, doing a bit of forensics on trade history..
Overall :great job Mate
 
peter2 said:
A little bit of psychology, a little bit of commonsense and a huge pat on the back. A trading system will only earn what it's designed to earn. Our role is to apply the system as perfectly as possible in order to earn the maximum that's available. If the system is not fully automated then we're going to make some mistakes. Overall, mistakes cost us and reduce system profit. Hopefully the system has enough of an edge that a few mistakes won't cost us too dearly. I don't like that you're only earning 1/2 of what your system should earn. That's a huge problem that you'll have to fix. The system could be at fault if the buys and sells are frequently impossible to get in the real market . Or you're the weak link in the system. You've got to minimise the number of mistakes.

I think you're correct in comparing your results against a system back test (or a forward test). Don't fall into the trap of comparing the results of your system against results of other systems. If you see that other systems return better results then work on developing better systems. @Skate is a master at this.

Don't overlook the significance of your trading achievements. Your MAP system started at exactly the worst possible time. Did you quit? No. You traded your system through a very difficult period and now you're starting to see the rewards as your equity curve leaves the XAO in it's dust. Very few new traders experience the immense personal satisfaction that you should be feeling now. Most of them quit or change systems. You have dug yourself out of a significant drawdown. Having done it once, you must know that you can do it again.
Click to expand...

Thanks mate. I don't think I've made too many mistakes overall, so I should look at the backtesting against my live results. I checked it a while ago and I was actually outperforming the backtest. Now it seems to be the other way. I think the few trades I've missed have made a difference (they have made a difference with the HappyCat strat as I'm behind skate with that due to the limited trades I missed). I'm not too worried right now as my backtest showed +60% which would be an absolute crazy amount. During my paper trading the backtest and trading matched perfectly. During the first 4months or so on live data, the same. So some investigation should probably be made.

You are right I started at possibly the worst time. It was hard to trade through all that as the market was crashing and my system hadn't given me sell signals. But what kind of system trader would I be if I discarded my rules immediately out of the gate? I also reminded myself that the best time to trade is after a market crash, but you can never pick the bottom, so you just have to keep trading and manage risk. It looks like that has paid off. The encouragement from users here has definitely helped. And as a trend trader I expect to be in a draw down in the future, but at least I know I will recover with a vengeance.
 
qldfrog said:
About the lag against backtest: i believe check against backtest is indeed the criteria to track.i stopped one of my system due to unexplained delta.
One way i have explained deltas which can seriously affect performance is when i have both buy/sell at open but no cash reserve left.
my broker does not accept buy order if i do not have cash or pending cash available, so i need to wait for some packets being sold before entering an extra buy order.
Based on sell execution orders at open, it sometimes mean my buy orders are late and miss the open price.
But 50pc underperformance is huge.worth investigating, doing a bit of forensics on trade history..
Overall :great job Mate
Click to expand...

Yea I think a few missed trades and not getting some fills may have left to the large difference in the delta. My back test runs on getting the next candles open. Though in practice I do the previous close +/-3%. I could simply change to just market orders at the open and not worry about restricting my limit price. I am lucky where I don't have to wait for settlement with my broker either, but as you mention if i don't get a sell order fill I have to wait for the buy usually (unless I have enough cash for the position).

I'll have to find some time to investigate and hopefully I can sort it out. +30% is nothing to baulk at though, especially since I was in a -23% drawdown. +60% would of course have been better ;).
 
I did a backtest from today back to september. Why september? that's when I added more capital to my system and the approximate time I also got out of a drawdown. It returns a profit of 34% which is damn close to what I'm at now. I think the issue may be the potential missed buys or sells when the market was a bit erratic/unsure. I'm also not sure if I could have handled buys/sells a bit better From Feb-Sep.

This brings me to a question for the followers. Should I just buy/sell at the open with a market order, or should I continue doing the +/-3% that I currently do? If the backtest just does the open for buys/sells perhaps to be true to the backtest I don't micromanage the orders and just let it enter at whatever the market is doing?
 
Warr87 said:
Should I just buy/sell at the open with a market order, or should I continue doing the +/-3% that I currently do? If the backtest just does the open for buys/sells perhaps to be true to the backtest I don't micromanage the orders and just let it enter at whatever the market is doing?
Click to expand...
@Warr87 how about modifying your code a bit so the backtest simulates your exploration - 3% limit as suggested by @captain black in the Dump it here thread then you might be able to understand your differences between the backtest and trading rules

Dump it Here

It's worth noting that Amibroker Backtest can't be overridden to compensate for missed (not purchased) positions. That's not correct. Use a buystop as part of your entry condition, adjust your trade delays and set the buystop entry condition to false if the low of the entry day gaps up...
www.aussiestockforums.com www.aussiestockforums.com

I found this code on the AB forum which works fine, you just have to remember to change your setting - SetTradeDelays(0, 0, 0, 0);


// I have added some comments to help

Code: 
/// based on limit order example of
/// @link http://www.amibroker.com/kb/2014/11/26/handling-limit-orders-in-the-backtester/
/// responded to in this thread
/// @link https://forum.amibroker.com/t/overriding-amibroker-backtester-settings-generating-unrealistic-results/7374/4
BuySignal = Ref(C, -1) >= 2.0; // Replace code ****Ref(C, -1) >= 2.0**** with your buy conditions

// buy on the next bar
Buy = Ref( BuySignal, -1);
BuyStopPrice = Ref( Close, -1) * 1.03; // 1.03 is 3 %

// now we check if buystop was hit
Buy = Buy AND Open> BuyStopPrice; // Was H is code and changed to Open

// if Open price is above the buystop, then we use Open for entry
BuyPrice = Max( Open, BuyStopPrice );
 
Warr87 said:
This brings me to a question for the followers. Should I just buy/sell at the open with a market order, or should I continue doing the +/-3% that I currently do? If the backtest just does the open for buys/sells perhaps to be true to the backtest I don't micromanage the orders and just let it enter at whatever the market is doing?
Click to expand...

Trav. said:
I found this code on the AB forum which works fine, you just have to remember to change your setting - SetTradeDelays(0, 0, 0, 0);
Click to expand...

@Warr87 I'm keen to hear what difference it makes on your backtests. For mine I get better results with just a Market Order as opposed to the C+3% rule.

@Trav. Great work on digging up that code
 
Trav. said:
@Warr87 how about modifying your code a bit so the backtest simulates your exploration - 3% limit as suggested by @captain black in the Dump it here thread then you might be able to understand your differences between the backtest and trading rules

Dump it Here

It's worth noting that Amibroker Backtest can't be overridden to compensate for missed (not purchased) positions. That's not correct. Use a buystop as part of your entry condition, adjust your trade delays and set the buystop entry condition to false if the low of the entry day gaps up...
www.aussiestockforums.com www.aussiestockforums.com

I found this code on the AB forum which works fine, you just have to remember to change your setting - SetTradeDelays(0, 0, 0, 0);


// I have added some comments to help

Code: 
/// based on limit order example of
/// @link http://www.amibroker.com/kb/2014/11/26/handling-limit-orders-in-the-backtester/
/// responded to in this thread
/// @link https://forum.amibroker.com/t/overriding-amibroker-backtester-settings-generating-unrealistic-results/7374/4
BuySignal = Ref(C, -1) >= 2.0; // Replace code ****Ref(C, -1) >= 2.0**** with your buy conditions

// buy on the next bar
Buy = Ref( BuySignal, -1);
BuyStopPrice = Ref( Close, -1) * 1.03; // 1.03 is 3 %

// now we check if buystop was hit
Buy = Buy AND Open> BuyStopPrice; // Was H is code and changed to Open

// if Open price is above the buystop, then we use Open for entry
BuyPrice = Max( Open, BuyStopPrice );
Click to expand...

Don't we only want to buy if the open is less than the BuyStopPrice?
 
I added this and I got 1 extra buy and no changes to my backtests lol.

Code: 
Buy = (MapBuyCondition AND PriceF AND Index_UP) AND Open < BuyStopPrice;
 
Top