Systematic Futures Trading

[jjbinks]

Hi All,

So as some may now my experiment trading US stocks using a trend following system ended last week. I decided the efforts compared to trading ASX did not justify taking it to real trading.

I am contemplating looking at developing and trading a futures system but keen to hear what experience others have had.

In terms of my own experience trading futures.
I have lost money trading futures in a discretionary way a few years ago. Despite this I still feel I will have a lot to learn about the intricacies of futures particularly if I want to build a system. I am doing reading to address this.

In terms of practicalities:
-I am thinking of trading longer time frame i.e. days rather than intraday
-I will be using norgate futures data
-I previously traded futures with AMP clearing but will have a look at interactive brokers. If I go with IB I could use their data.

Time frame:
-Next 2 months will be busy for me with my job but I intend to read and solidify my knowledge
-Late Nov/Dec I may get data and start testing ideas
-Trade dummy account sometime early next year and then take things from there?

Would love to hear comments, questions, suggestions or if you think it is a stupid idea!

Cheers
JJ

 

[kahuna1]

Hmm ...

Well trading is something I have spent my life at. Nowadays as you seem to be aware that most trading is done via computer systems verses the human decisions.

As a fundamental sort of a guy, one who reacts to news or forces, in this modern world its a loose loose game at best. bad news comes out. Terrible news, market tanks, as expected, the humans do the correct thing, then the computer, idiot savant trend following imbecile machines see the selling has abated, sense there is more risk to the upside and hey presto ..... despite something that will doom say a stock totally, or an index or commodity ... or currency ... send it say a 12 months range one way over time, the idiot savant machines just follow the trend.

So the sell off stops ... its gone down 5% of whats likely a 15% move, but for now the selling has stopped, slowly it creeps up and up and UP and those who missed the initial sell ... which is virtually everyone ... so they sold 2-3% down ... the commodity or index goes from 5% down to 4% to 3% ,,, basically break even for most, then its down a mere 1% ... and all hell breaks loose as the idiot savants who followed on its downward trajectory ... dont CARE ... that say the news of shocking proportions longer term occurred and off go the SHORTS stop losses ... and its back to unchanged .... then up 1% up 2% ... up 3% ... ab0ve the bloody starting point.

Systems work fine .... SHORT term ones .... technical ones ... BUT they require diversification and trading say 100 comm0dites ... index's or stocks and currencies or currency crosses. Basically at any one time about 20 of the parameters in the 100 things your trading will either NOT be working or totally NOT working for about 10 .... so one needs diversification and massive computing and blinding speed to arbitrage at times things that seem not so connected but totally are such as say OIL to an OIL stock being traded and if its via a USA repository .... then comes in the exchange rate into the mix ...

I could go on .... having built and run such things for 20 plus years.
Thinking there is some magical or mythical system that will work all the time or even close to it, is .... a dream.

There is no such thing.
Discipline and a computer driven model is taking any human decision OUT of your hands. At times even that will NOT protect you in times of extreme news ... or say markets loosing all liquidity and even worse hitting limits for maximum daily moves and then opening the next day without trading locked limit UP or LIMIT down ...

Bottom line, get say 30 quant guys and great programmers, about 100 million in capital and then one needs to be a broker and have execution rates of say $1- per futures contract or about $1- per $50,000 of stock so that's about 20% of what most pay for stock trading and 10% for most futures guys and YOU may stand a chance.

Sorry, but ... being honest.
Reading some fable tellers book about trading and follow this indicator or that one, has made many people rich and none of them via following some secret system. Some things work, then dont work, most wider longer term things just no longer work because RISK in the times we are in is not respected and with no human going, hey didn't that building that just had its stock appreciate 20% burn down ? Is quite a common occurrence.

Quite often, and amusingly so. In say the year 2000 ... when I still had no grey hairs, the most admired company was a stock called Enron .... which amusingly on the same list which Google has erased from searches ... it was the Fortune Magazine list, had the least favored stock or company one called Trump Casino resorts.

Amusing now in 2019, Enron which was mixed up with George W Bush Senior and Junior ... heavily involved in them, exposed as a sham and bankrupt ... loosing 3 billion of employees super funds and about 8 billion more and now in 2019 ... we have the fooool who was a crook ... in charge.

Good luck with things. Diversify .... SHORT term .... High frequency ... it does work ... but you need 100 if not 500 things you trade or arbitrage so to speak.

 

[aus_trader]

Wish I could provide actual experience, but I am no expert in Futures. Have traded Forex for a period of time but it's actually quite hard to make money consistently in Forex. Not to discourage you but I assume it may be as difficult to do so in Futures ? But since you have some past experience, you are in a better position to decide. Keep us posted with your progress...

 

[jjbinks]

@kahuna1
Appreciate your honesty. Its what I'm after.
I think some of what I have read so far echo's your message although the books are not so explicity/gloomy (they can't be if they want to sell!). But may strategies need require very diverse portfolio which needs much more capital than you need to trade stocks.

 

[kahuna1]

Appreciate your honesty. Its what I'm after.
I think some of what I have read so far echo's your message although the books are not so explicity/gloomy (they can't be if they want to sell!). But may strategies need require very diverse portfolio which needs much more capital than you need to trade stocks.

Sorry ...
If there was a secret system .... NO ONE ... would sell it.
There are literally thousands of math masters, quant guys ... coupled with the best computer techs and artificial trend following programs that do this every day.

YES ... it works overall, works quite well ... but unless you have ultra low cost brokerage and diversification over 100 plus products, it will not and does not work. Yes you may find the perfect system that works when back-tested and makes 10,000% over the last 12 months, likely it will break even at best the next 12 month or if you use 20 years back-testing you will find its the worst system EVER to use some years.

Reality is what it is.
I do find tech stuff of value ... when coupled with a very firm fundamental view. Quite often, the idiot savant models drive prices to absurdity prior to crashing. Then again, under estimating stupidity is never a wise choice. At times, shares worthless one can clearly see go from 10 cents to $10- and whilst they may eventually go back the whole way down, selling into ... a fools rally is why one diversifies and ant times MUST have extreme discipline and a stop loss.

No magic wisdom sadly from me.
The sellers of books or sage advice make more from selling courses ... books or rubbish than any advice they likely ever will give you.

That is unless is meat and potatoes common sense stuff. Stop loss .... diversify risk. Sell into rallies and buy into dips. NOT the opposite.

 

[Joe90]

For some insight into futures trading suggest to attend the church of Minwa.
https://www.aussiestockforums.com/posts/897305/
Take notes, copy everything of value into a word file, print, digest regularly, with plenty of demo account practice.
See this excellent post by Trembling Hand on learning via fast track demo trading.
https://www.aussiestockforums.com/threads/the-transition-to-futures-trading.26509/page-2#post-763039

 

[jjbinks]

@Joe90 those are great resources.

 

[Newt]

Watching this new foray with interest jjbinks

 

[jjbinks]

just an update. finding a daily system is not so easy. (For reasons mentioned above).
I have also been very time limited but now as I will be mostly working from home I may have some extra time.

Current market conditions highlight importance of have diversification in systems (assets) in your portfolio. Basically momentum with stocks is not going to be pretty with current volatility and big falls. But the recovery will come!
Obviously if you had short/long system you might be bit better with stocks.

I've had a few lucky discretionary trades on SPI/ES futures. I guess its similar to the story of being easy to make money in bull market with stocks (going long). In current climate short trades can be rewarding but also risk as you don't know when the tide will turn.

Just FYI
Over last 12 months I was trading two systems.

1) Based on tech trader as described in holy grail. As of two weeks ago was running for 52 weeks. I was thinking of getting stopping the system at the 40 week mark and am now out completely.


2) Trend system based on linear reg (from Clenows book trend following with stocks)



Only positive is that System 1 I stopped basically first week of march and system 2 has had now new trades since 1st week march.

Hope everyone is going OK!

Cheers
jj

 

[qldfrog]

Hi

just an update. finding a daily system is not so easy. (For reasons mentioned above).
I have also been very time limited but now as I will be mostly working from home I may have some extra time.

Current market conditions highlight importance of have diversification in systems (assets) in your portfolio. Basically momentum with stocks is not going to be pretty with current volatility and big falls. But the recovery will come!
Obviously if you had short/long system you might be bit better with stocks.

I've had a few lucky discretionary trades on SPI/ES futures. I guess its similar to the story of being easy to make money in bull market with stocks (going long). In current climate short trades can be rewarding but also risk as you don't know when the tide will turn.

Just FYI
Over last 12 months I was trading two systems.

1) Based on tech trader as described in holy grail. As of two weeks ago was running for 52 weeks. I was thinking of getting stopping the system at the 40 week mark and am now out completely.
View attachment 101567

2) Trend system based on linear reg (from Clenows book trend following with stocks)
View attachment 101569


Only positive is that System 1 I stopped basically first week of march and system 2 has had now new trades since 1st week march.

Hope everyone is going OK!

Cheers
jj

Jj, on your first graph, xao is still positive after a year, ???

 

[jjbinks]

@qldfrog graphs are based on data from when i stopped trading 2 weeks ago.

 

[qldfrog]

@qldfrog graphs are based on data from when i stopped trading 2 weeks ago.

ok
.thanks

 

[KevinBB]

I know this is an old thread, but there isn't much on television ...

If you are still interested in developing your own automated futures trading system, read "Systematic Trading" by Robert Carver.

Carver presents a framework for trading futures on a medium to long-term basis, together with a basic system that uses that trading framework. If you want short-term or day trading, Carver's framework is not for you.

The system involves using multiple moving averages, multiple breakouts (they aren't really breakouts), and a carry rule. The system is just there to demonstrate how the framework works, and it is relatively easy to implement your own system and bring it into the framework. The framework isn't only about futures, it can be adjusted to trade stocks as well.

I started out about one year ago by implementing the framework and system in a series of spreadsheets, using index and commodity CFDs for one system, ASX 20 stocks for a separate system, and also US Sector Indices in third system. All three worked well, but that could just have been because we have had a very strong trending market in the past year or so. Moving averages and breakouts work very well in trending markets.

My system now has progressed to auto trading a limited number of US futures during the Chicago day trading session, using Carver's framework inside a self-written Python script. I sleep all night (Sydney time) while this is happening. Learning Python isn't necessary, initially I would implement the framework on a spreadsheet, and manually trade securities (futures, stocks or CFDs) in a timeframe with which you are comfortable.

Carver has written other books, too. If you're interested, read "Leveraged Trading" first, then "Systematic Trading".
PS: I'm not getting paid to write this ... Carver is an ex AHL systems man, blogs about his system and experiences, and trades something very similar to the framework described in "Systmatic Trading" for his own investments. His system, spreadsheets and Python code are published for all to use, but I would still start by developing my own spreadsheets.

KH

 

[Warr87]

I'm a fan of Carver and his books.

 

[bsnews]

I would like to move into system trading or at least learn how but I just can't get my head around the programming of Amitrader or RealTest. I am good with spreadsheets though, would these books shed some light on the programming side?
Cheers

 

[KevinBB]

Hello @bsnews

The books won't teach you how to program. What they do is to provide what Carver calls a "Framework", something for you to build or to adapt. It includes things such as using volatility for position sizing, and handling transactions that aren't based in AUD.

Carver uses Python to code his systems. I am an experienced computer programmer, but of the ancient (pre 1990s) style, i.e. I don't use the modern OOP or classes. So, I didn't understand one word of Carver's Python code, well, didn't understand most of it is more accurate. What I did was to put the framework into spreadsheet format, a format which I understand and a format for which I was able to verify results. Stick to what you know.

Carver has an extensive blog (here) and website (here) and his full Python code (here), but it can get quite technical. I have a good knowledge of mathematics and number, but his statistics knowledge is far above mine. So, I stuck to the framework in the books. The website has a resources section, which includes sample spreadsheets. I didn't use his spreadsheet samples as a final system, they are quite basic, but instead used them to verify my own calculations.

From books to a fully automated live trading system took me about 10 months or so, almost full time. I'm a slow learner (not really). Started with spreadsheets trading CFDs to get the diversification, then progressed to futures (I do have futures trading experience in a past life) and US Sector ETFs and AU top 20 stocks, still with spreadsheets. It is only in the last 3 months or so that I devoted time to learning Python, and adapting my spreadsheets to Python code.

If you are going down the Carver path, start with Leveraged Trading, then progress to Systematic Trading. I used Leveraged Trading to set up the first CFD spreadsheets. Its a bit easier to understand the logic and framework in this book, or was for me. Systematic Trading goes to the next level. The book on portfolio management is another step up again.

Hope this helps a bit. I'll have to hit him up for commission!

KH

 

[Warr87]

I agree with @KevinBB on all of that. I have also found some traders that have gone from Excel to python, so trading in Excel is definitely possible.

Amibroker and RT can be a bit hard, particularly if you aren't used to coding. If you keep at it, you will find it will eventually click. Carver's code (or rather his pysystemtrade) is also not user friendly but it is available. he uses it for himself but if he has issues he obviously fixes it. i haven't kept up with his website for about a year but it is worth a read. him giving the code away for free and being upfront with it is definitely not something he needs to do.

Also, I found tinkering with code is a great way to learn. And google is your friend. I also ask people who are much better than me when I cna't figure it out myself (last resort but it is needed). So don't be discouraged. Even top-level programmers still google on how to overcome programming errors.