Australian (ASX) Stock Market Forum

Weekly Portfolio - ASX

Jump to new
Week 9:

Buy: ECX, MAH
Sell: HT1

Current Portfolio
RBL Bought @ 1.395 (12/8/19). SL 1.10. Current uPL: 0.10R
WSA Bought @ 2.48 (12/8/19). SL 2.36. Current uPL: 0.87R
SCP Bought @ 2.56 (26/8/19). SL 1.84. Current uPL: 0.09R
IFN Bought @ 0.495 (26/8/19). SL 0.47. Current uPL: 1.15R
CDD Bought @ 1.10 (23/9/19). SL 0.77. Current uPL: -0.07R
HT1 Bought @ 1.875 (23/9/19). SL 1.312. Current PL: -0.14R
ECX To be bought at 30/9/19 open.
MAH To be bought at 30/9/19 open.

WPP already closed out for a -0.50R loss.

Current unrealised PL: 1.5R

I have hopes for RBL, WSA, IFN & ECX. With HT1 being closed out I'll have a realised loss of -0.64R. Obviously a pretty small amount and I am happy to close out losers quick. Though HT1 only lasted 1 week. Don't want a whipsaw going on there.
 
Week 10:
Buy: No buy signals
Sell: MAH

Down a little this week with MAH being closed out and the sell for HT1 my unrealised PL is 0.98. WSA and IFN are keeping me in the positive right now. Trying not to get hung up on numbers obvious as being so early into this I will have more losers than winners. My win% should be between 30-35% based on backtesting.
 
Week 12:
Buy signals: EHE, NWH
Sell signals: SCG

RBL - 0.25R
WSA - 0.70R
IFN - 1.22R
SCP - 0.15R (To be closed on Monday)
CDD - (0.53)
ECX - (0.20)
EHE & NWH to be bought on Monday

Total right now: 0.80R
 
Update:
When I went over my buy/sell signals I noticed a mistake and couldn't edit my post for some reason. I wont be closing out SCP.

I also was curious and ran my code through the backtest and found that I would have nothing but losses from the time I ran this paper trading. I'm not sure why I have some buy signals when I run this system weekly but not all of them show up when running it historically over the past few months. Either way, my paper trade is at least positive. Another interesting fact is that if I had started this system in February then I would be doing very well as there were a couple of big buys in Feb and April. As a lot of us know, timing means a lot!

I have also been reading some books and going back over some posts. While I am testing this system I am still interested in testing some other systems as I might even start trading in a SMSF (which will have a lot more capital then what I can get in my personal account right now).

I finished reading "Trading Evolved: Anyone can build killer trading strategies in python" by Clenow. I enjoyed that book a lot (which is probably why I read it in a few days despite it being 400 pages). I'm trying to learn python so this was a good book for me. The book also points out that it is better to have a combination of strategies (stocks, futures, etc.) running at one time to be the most effective. Though .... I don't have a few mil to do that right now ;).

I also read Nick Radge's "Weekend Trend Trader" last week.

The book I'm reading now is "A Complete Guide to Volume Price Analysis" by Coulling. I'm looking forward to this as I don't use volume much in my trading but I think it would help a lot.

I am very much open to any other books (preferably on Kindle/Amazon) that people think would be useful. I will likely work towards paper trading a few different systems. I am also very much towards the systematic trader route (would even like to automate my systems at some point too).
 
Checkout other books by Andreas Clenow.

His latest book Trading Evolved is definitely worth the $8-9 it is on Amazon.
Keep working on learning python. If you are interested in learning a bit more about using zipline package check out tutorials by sentdex on youtube. He also has good general python tutorials but you have many choices for this.

Not sure if you use norgate data but if so they are alpha testing a python plug in and a zipline plug in. I've been playing around with it and looks promising. The zipline plug in is based on Clenow's scripts in his book.
 
Thanks @jjbinks . After reading his book I definitely felt that his other books would be good too. And youbarr right, for how cheap it is on Amazon, it's amazing.

I bought historical data from premium data a couple of months ago. I'm tempted to actually sign up with them for a monthly subscription. I know norgate is another good supplier. The fact they are getting those plugins is great (And I wish I knew that before buying historical data from premium data). I've been wondering how I can export my data from amibroker for zipline. I have to look at this on the weekend but I think it's possible to just export as a csv and then tell python/zipline the columns. Things to look in too!
 
Warr87 said:
Thanks @jjbinks . After reading his book I definitely felt that his other books would be good too. And youbarr right, for how cheap it is on Amazon, it's amazing.

I bought historical data from premium data a couple of months ago. I'm tempted to actually sign up with them for a monthly subscription. I know norgate is another good supplier. The fact they are getting those plugins is great (And I wish I knew that before buying historical data from premium data). I've been wondering how I can export my data from amibroker for zipline. I have to look at this on the weekend but I think it's possible to just export as a csv and then tell python/zipline the columns. Things to look in too!
Click to expand...
norgate and premium data are same supplier just slightly different format but same data.
If you are thinking of getting premium data subscrption I would recommend just going with norgate so you could register for alpha testing.
The only downside is that you don't get to keep the historical data after the subscription ends.
You definitely can work with python and csv from premium data. In fact I did that a few years ago. But it does require a bit more programming. With the python-norgate plug in a lot of the work making data the right format is done for you. Also it is more convenient if you plan to test historical index constituents.
You still need to be OK with python basics even with the plug ins.
 
Hey everyone. I bit the bullet and finally got Premium Data. I've already converted my data to CSV files for any python work I do later. Though I am busy working on other things right now (but have 10 years of data ready to go).

I am still reading the Volume Price Analysis book. It's not too bad, though I have found it a little repetitive. I will reserve my overall opinion until I finish the book.

I have been in conversation with @Skate who has been kind enough to help out. Soon (either today or hopefully Monday) I will start paper trading another system. It will still include an EMA close but will actually be my version of @Skate's MAP system. Back tests have been good, what I'm lacking is consistency in my backtests.

Week 13:
Buy: ACR
Sell: CDD (this nose dived. Lost 40%)

RBL - 0.68R
WSA - 1.20R
IFN - 1.02R
SCP - 0.25R
CDD - (1.63)
ECX - (0.20)
EHE - 1.94R
NWH (0.01)

As you can see, CDD is going to errode what should have been a good week. This happens. Hopefully CDD doesn't slipe any more at Monday's open. And my new buy, EHE has done very well in it's first week!
 
Week 14:
Buy: no buy signals
Sell: NWH

My overall P/L is $1,007.55 so far. Biggest winner so far is EHE, then WSA and IFN.
 
System 2 — MAP and EMA Strategy

A MAP inspired strategy to be paper traded (to test its viability as a secondary system on a SMSF). The system is still be coded to improve consistency and will be adjusted as it is evaluated with backtesting.

Week 1:
Buy: ERA,LNG,PAN,SWM,FLC,ISU,DNK,PRU,DCG,SLR,MVF,TGP,OFX.
 
Some clarification. My P/L of $1007 is from my starting capital of only $20,000. That includes commissions as well. So while it isn't a lot, it's not too bad 14 weeks in. I'm not even maxed out in my positions yet.

System 2 Trade Plan:
Goal - to be consistently profitable trader. Aiming for 15% P.A. with max DD of 20%.
Entry/setups - Based on MAP strategy of Skate's.
Scans- Using AB to scan market EOW. Data supply from Premium Data.
Money management / Risk management - Starting capital is $50,000. Max 20 positions with 40% trailing stop or 10% trailing stop (market up = 40% trailing stop, market down = 10%) puts risk at 2%. Once capital has grown and position size reaches $15,000, I will look at increasing max positions for $15,000 parcels.
Exits - Exit is either triggered via trailing stop, stale exit, or EMA cross. Assessed EOW.
Other - Going long only. System will perform better in an overall upwards trending market.
Universe - All Ords

Criticisms, notes, remarks are welcome! This is being paper traded for a possible use on either personal account or creating a SMSF (if a SMSF, starting capital will be higher). Backtesting has been completed over the past three years. I am still tweaking the system to get greater consistency though overall system has performed very well.

Thanks to @Skate for providing the MAP code buy signal!
 
System 2: I have added in a price filter and some out of sampling optimization showed some good results. My defaults were relatively good and I have tried to avoid over-fitting. I got an average of 30% annual returns (2018 continues to be a draw on my system).

What I want to look into now is using the PositionScore variable better. My system can often find buy signals and I think with better scoring and possible volume filter, I can improve my chances.
 
Positionscore is also another opportunity to align your system picks with the sort of trades you prefer to take. Another approch might be to score low trades you'd struggle to reliably take. Assuming you have experience to recognise what would be hard to trade. Of course whatever scoring code you use it should stand up to robust backtesting too :2twocents
 
Top