Can mechanical forex systems work?

I've got to agree with littlemog. Reading about (some of) the experiences of two people that are many paces down a path I'm starting to follow is very interesting. I've learned a few things too.
 
When I get back to testing systems in a more thoughrough and systematic way I'll post my findings in my journal. I'm expecting to find EoD systems that work but they'll probably use very different methods of analysis to any (successful) shorter term systems I come up with.

Definitely going to explore both avenues though. If I can develop EoD algorithms that are as successful as my shorter time frame ones I'd rather go with EoD and save on the headache of getting a system fully automated.

Mind you I'd probably try to automate it at some point anyway just for the challenge:whistling

If you can get EoD working in a strategy, I think that's a fantastic plan. One of the major MAJOR issues with automated strategies is the spread, slippage and commission on each trade. If you're working on EoD, that means you're placing a fraction of the number of trades, which in turn means that these factors should become insignificant. So far, I've got my system down to trading every 6 hours or so, on average, but if you can get decent returns on once every few days automation would be almost trivial.
 
Thank you north5 & rnicoll for this very insightful thread.... even though I don't trade with systems, it's a good thread to follow and I've learnt much. Keep it coming!

Thank you (and Vrothdar), it's good to know we're being useful :) I've just been asked if I blog about work on a project I'm doing in my free time (work related), and it occurs I may throw together a blog for this, too. Oh, I'm also intending to put up a $50/month system feed on Collective 2, fed out of my order confirmation step :) So if people are interested in trying to trade the same signals, and can accept they'll be half a second behind me, that will be an option.

In the short term, however, I see a lot of rest in my future. I've been doing what amounts to 90 hour weeks for about two months now, and it's really beginning to show.
 
Blog you say? Yes, that'll be very nice, :clap::D

Thank you (and Vrothdar), it's good to know we're being useful :) I've just been asked if I blog about work on a project I'm doing in my free time (work related), and it occurs I may throw together a blog for this, too. Oh, I'm also intending to put up a $50/month system feed on Collective 2, fed out of my order confirmation step :) So if people are interested in trying to trade the same signals, and can accept they'll be half a second behind me, that will be an option.

In the short term, however, I see a lot of rest in my future. I've been doing what amounts to 90 hour weeks for about two months now, and it's really beginning to show.
 
Blog you say? Yes, that'll be very nice, :clap::D

Working on it :)

I've thrown the system onto Collective 2 ( Catching Daggers - C2 trading system which trades forex ), and will manually copy trades onto Collective 2 for a bit if people want to see them. Live trades are subscription-only, but you should be able to see trades once they close, for free.

Can I get some feedback on whether people are interested in autotrading this? As it stands, Collective 2 is fairly bloody expensive for something I thought I'd do for a curiousity, and if people aren't too interested in using their platform specifically I'll probably have my system generate an RSS feed of orders instead. On the other hand, if there's enough people who'd like to trade along with me to actually cover the costs of Collective 2, that's another matter.
 
Working on it :)

I've thrown the system onto Collective 2 ( Catching Daggers - C2 trading system which trades forex ), and will manually copy trades onto Collective 2 for a bit if people want to see them. Live trades are subscription-only, but you should be able to see trades once they close, for free.

Can I get some feedback on whether people are interested in autotrading this? As it stands, Collective 2 is fairly bloody expensive for something I thought I'd do for a curiousity, and if people aren't too interested in using their platform specifically I'll probably have my system generate an RSS feed of orders instead. On the other hand, if there's enough people who'd like to trade along with me to actually cover the costs of Collective 2, that's another matter.

correct me if I'm wrong, but C2 take a about a third of your subscription. Other than that, it costs about $60 for a 6 month listing, so your costs don't need to be covered, do they?


UTB
 
correct me if I'm wrong, but C2 take a about a third of your subscription. Other than that, it costs about $60 for a 6 month listing, so your costs don't need to be covered, do they?


UTB

$60 for 6 months? I may have read the details while sleepy, and thought it was $60/month.

Still, if people just want to see the trades, and don't care about the platform, I can throw an RSS feed onto a website for free. Although I suppose it's probably better to start having a record of trades on Collective2 so that even if I don't want to get heavily into advertising it now, I have the option later...
 
$60 for 6 months? I may have read the details while sleepy, and thought it was $60/month.

Still, if people just want to see the trades, and don't care about the platform, I can throw an RSS feed onto a website for free. Although I suppose it's probably better to start having a record of trades on Collective2 so that even if I don't want to get heavily into advertising it now, I have the option later...

I'm sure it's much less than you're thinking. The chunk of subscription fee is large mind - but that is a problem you will want to have:D

The track record you will build there will be the advertising you will need - so it's worth it IMHO.

Good luck,
UTB
 
rnicoll,

sounds good... would love to see the trades in action (RSS or otherwise). Wouldn't mind a pm when your blog is up... :)
 
OK - spent quite a few hours over the weekend having look at MetaTrader. don't think I could get it to do what I want. It appears to be entirely indicator driven - There isn't really any way to record paterns like I do in my DB without a huge series of Arrays or something.

So basically I wasted a whole load of time and achieved nothing.


OK I have added an extra exit strategies to my program so I can evalute if 20:30 or 50:30 works best, and tidied the screen up a little - so I have moved forward a bit. Would like to try and work out how I can implement a trailing stop next I think.

I also looked a C2 - got the code to send my trades to it commented out at the minute. Didn't fancy paying to prove it works until I'm a bit more sure it will.
 
Last edited:
So basically I wasted a whole load of time and achieved nothing.

You made the achievement of being able to say, with certainty, that metatrader can't easily do what you want.

Sounds like things are moving on nicely though. Hopefully by next weekend I'll have a copy of excel 07 and will be able to spend some serious time trying to clean up the tick data I'm in the process of downloading.
 
You made the achievement of being able to say, with certainty, that metatrader can't easily do what you want.

Yup. Never underestimate the usefulness of discovering how much most of what's out there is brain dead :) It's why I still read the news...

Sounds like things are moving on nicely though. Hopefully by next weekend I'll have a copy of excel 07 and will be able to spend some serious time trying to clean up the tick data I'm in the process of downloading.

Sounds good, best of luck, and do tell us how it goes, yeah?
 
Sounds good, best of luck, and do tell us how it goes, yeah?

Aye, will do. Got a few things I need to work out first though. Will make a thread tomorrow asking a few questions to try and speed the process up. For now though - bed.
 
Will PM you in a second, but for everyone else, it's going up at:

Fox Quant Blog Archive Introduction

(Fox Quant is the name my software will have if I ever release it officially)

Maybe not bed just yet - just had a read through your blog so far and it seems quite interesting, keep up the good work.

I really wish I had the time to learn a bit more about about computer technology and get myself to the sort of level that I could use linux with a reasonable degree of competence. Unless I end up down the road of intraday trading and therefore need to automate I can't see it being particularly worthwhile; at least not while I'm working full time.

For now my ability to use VBA/excel is doing enough for me. I may get the opportunity to learn SQL at work which I expect might help as well.

The only thing I'll say about data feeds is this: Don't get me started!
 
Maybe not bed just yet - just had a read through your blog so far and it seems quite interesting, keep up the good work.

I really wish I had the time to learn a bit more about about computer technology and get myself to the sort of level that I could use linux with a reasonable degree of competence. Unless I end up down the road of intraday trading and therefore need to automate I can't see it being particularly worthwhile; at least not while I'm working full time.

Don't be too afraid of Linux, it's actually quite approachable these days, especially for simple tasks like e-mail/web browsing. Ubuntu Home Page | Ubuntu is a really good distro for the novice user, and they do a CD you can boot from and it shows you what Linux looks like (although running from CD obviously it's rather slow).

For now my ability to use VBA/excel is doing enough for me. I may get the opportunity to learn SQL at work which I expect might help as well.

The only thing I'll say about data feeds is this: Don't get me started!

If you don't know SQL already, I wouldn't stress too much over it. It takes a while to learn, it might well be the case that you can't save enough time using it to make the investment worthwhile. That said, my platform's available as an example (PM me when you're about ready to start coding, though, as it changes fast so there's no point in me sending you a version that will be several weeks out of date by the time you get to looking at it), and may give you some help with SQL.

Datafeeds; yeah. I've dealt with a few pro feeds, too, as part of work, and they're generally not any better. It's despair-inducing. Actually seriously considering looking into starting a data feed service; NxCore squishes data a little smaller than I've figured out how to do, but I reckon I can stream 500-odd shares at 5 second bars for 8kb/s (or about one thousandth of a good home network connection), with better accuracy and arrival times than almost all of the feeds out there.


Tonight should be interesting. The test version has been modified to try recovering when IB reboot all their systems at 5am (UK time). I expect it to explode so hard they're picking pieces out of the moon, but it would be nice if it actually manages it.
Tomorrow's looking like a good trading day, for me. No significant news coming out on the currencies I trade (and news does wreck my system)...

Which probably means I'll be broke by tea time, thankyou Murphy.
 
Don't be too afraid of Linux, it's actually quite approachable these days, especially for simple tasks like e-mail/web browsing. Ubuntu Home Page | Ubuntu is a really good distro for the novice user, and they do a CD you can boot from and it shows you what Linux looks like (although running from CD obviously it's rather slow).

I was thinking of trying ubuntu quite a while a go. Will probably give it a whirl when I get around to building a new machine at home. That'll be in a few months time at best:confused:

If you don't know SQL already, I wouldn't stress too much over it. It takes a while to learn, it might well be the case that you can't save enough time using it to make the investment worthwhile. That said, my platform's available as an example (PM me when you're about ready to start coding, though, as it changes fast so there's no point in me sending you a version that will be several weeks out of date by the time you get to looking at it), and may give you some help with SQL.

It's not something I'll bother with unless I learn it through work. Being paid to learn it would probably make the time investment much more worthwhile! With any luck I'll start on it by the end of the year - as such I wont be relying on it for trading related stuff!

Datafeeds; yeah. I've dealt with a few pro feeds, too, as part of work, and they're generally not any better. It's despair-inducing. Actually seriously considering looking into starting a data feed service; NxCore squishes data a little smaller than I've figured out how to do, but I reckon I can stream 500-odd shares at 5 second bars for 8kb/s (or about one thousandth of a good home network connection), with better accuracy and arrival times than almost all of the feeds out there.

If this ever happens let me know! How did the IB system reboot go? Has your system inhabited the moon?
 
It's not something I'll bother with unless I learn it through work. Being paid to learn it would probably make the time investment much more worthwhile! With any luck I'll start on it by the end of the year - as such I wont be relying on it for trading related stuff!

I suspect that yes, being paid would make it worthwhile :D Any idea what database engine they're going to be teaching you? Oracle, MySQL, etc.?


How did the IB system reboot go? Has your system inhabited the moon?

Completely failed. The restart code triggered, and it didn't work. Their stuff's fault, but it would still be nice to have a workaround. Especially as my system made 30 pips on GBP/USD overnight... on the paper version... I start up the live version this morning, and it loses 4 pips, or about half what I made on Friday.
 
I suspect that yes, being paid would make it worthwhile :D Any idea what database engine they're going to be teaching you? Oracle, MySQL, etc.?

Not a clue.

Completely failed. The restart code triggered, and it didn't work. Their stuff's fault, but it would still be nice to have a workaround. Especially as my system made 30 pips on GBP/USD overnight... on the paper version... I start up the live version this morning, and it loses 4 pips, or about half what I made on Friday.

That sucks. It's encouraging that your system's paper results appear to be pretty good, just a shame it's proving difficult to get those results with actual trades at the moment.
 
That sucks. It's encouraging that your system's paper results appear to be pretty good, just a shame it's proving difficult to get those results with actual trades at the moment.

It may be the best thing for me to do is put stop/limit and a timed stop on the trades, and accept that it'll be closing out automatically rather than guided, overnight.

I should mention, by on paper I mean that it's on a paper trading account with my broker, so these are realistic fills :)
 
Top