Systematic trading | Page 49 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Trade another day through Systematic Trading 203 replies

Nonparametric Association Measures in Systematic Trading 57 replies

Anyone use Systematic from Alpari? 1 reply

A Systematic Approach to Markets 68 replies

Systematic Trading vs. Discretionary Trading 14 replies

Attachments: Systematic trading
Exit Attachments

Systematic trading

  • #961
  • Jun 11, 2010 7:30pm Jun 11, 2010 7:30pm
http://www.nuclearphynance.com/Show%...stIDKey=143031
The breaking of a wave cannot explain the whole sea.
  • #962
  • Jun 11, 2010 8:43pm Jun 11, 2010 8:43pm
Well, this was one of the most informative things I've read about HFT. Which shows that there really isn't anything concrete written about it at all

Gave me an idea to test also.
  • #963
  • Edited 11:51am Jun 12, 2010 8:45am | Edited 11:51am
Sometimes NP is a funny place...first they bitch about him sounding like an idiot...and then they bitch about him leaking to much important info...

I found the reddit thread quite interesting.

Here is another piece of info:

http://wallstreetandtech.com/trading...reettech_daily

Seems everyone and their mother wants to do HFT FX.
  • #964
  • Jun 12, 2010 10:07am Jun 12, 2010 10:07am
Interesting that thay are also using the FXCM "bucketshop" (FXCM Pro maybe?) for liquidity access:
Quote
Because spot currencies are traded on multiple venues rather than on an exchange, CalPERS uses a number of electronic-trading platforms to access the liquidity of various banks: LavaFX (acquired by FXall), Credit Suisse Advanced Execution Services (AES), FXCM and Barclays Capital’s platform.

....

By using these platforms, you will have access to various pools of liquidity and not just banks — you might get a better price.
  • #965
  • Jun 12, 2010 10:37am Jun 12, 2010 10:37am
http://www.technologyreview.com/blog/guest/25308/

Quote
The five verbs with highest negative impact on stock price are hereto, comparable, charge, summit and green. If the verb hereto were to appear in a financial article, AZFinText would discount the price by $0.0029. While this movement may not appear to be much, the continued usage of negative verbs is additive.

The five verbs with the highest positive impact on stock prices are planted, announcing, front, smaller and crude.
  • #966
  • Jun 12, 2010 12:04pm Jun 12, 2010 12:04pm
Quoting mikkom
Hello all,

I decided to open this thread to discuss about systematic trading and development of strategies in "quantifiable" way.

At this moment I have decided to move away from metatrader that I have primarily been using for my trading strategy testing and development [silly, yes I know]. I have been reviewing all kinds of platforms and all of them seem quite not what I'm looking so I'm going to build my own testing platform during the thread and discuss the process [more about it during the thread].

I'm going to go a bit different...
This is a nice one i hope to gain from this.
  • #967
  • Jun 12, 2010 3:37pm Jun 12, 2010 3:37pm
Quoting BlackMage
Sometimes NP is a funny place...first they bitch about him sounding like an idiot...and then they bitch about him leaking to much important info...
LOL, I think the main theme is just bitching.
The breaking of a wave cannot explain the whole sea.
  • #968
  • Jun 12, 2010 6:52pm Jun 12, 2010 6:52pm
Quoting Craig
http://www.nuclearphynance.com/Show%...stIDKey=143031
Darn. Reading the more technical sections of that forum really makes me feel like a monkey on a typewriter.

Still, all this stuff is useful, if only to google the crap out of it for now.
For example, on the reddit post, one of the questions mentions kdb+, which once googled points to an interesting DB specialized for tick storage, complete with diagrams, feature list, trial software and code sample/support site. That's the kind of stuff that can help a lot with one's own design.

Or you have stuff like http://www.nuclearphynance.com/Library/, which has links to two PDFs roughly describing the knowledge domains required to get a job as a junior quant, which is pretty educational for a complete outsider to those fields, assuming those requirements map approximately to the minimum amount of knowledge required to be competent in that area.
How did this get here oh god I am not good with computer.
  • #969
  • Jun 16, 2010 6:00am Jun 16, 2010 6:00am
Quoting mikkom
There are some links to open source platforms on the end of the post #1 of this thread.
Thanks Ive downloaded and installed the marketcetera platform on my VPS.

How do you add forex data feeds to it I can't seem to find the option.
The greatest force on earth is compound interest.
  • #970
  • Jun 16, 2010 6:12am Jun 16, 2010 6:12am
Quoting krakoukas
I'm roughly in the same boat as you are, in that I'm building up my own little system from scratch, and trying to reuse as many open-source chunks as possible to make it easier.

Regarding the dukascopy tick-data, I'd recommend to not convert them to CSV files, as it's both much larger on disk and slower to use.
Stick to the download php script, although I'd suggest unzipping the files you download to make reading them faster. Or you could just leave them alone for now and unzip them on first read.

I bumped into this page http://ta-lib.org/hdr_lnk.html...
Whats an ATS?

It mentions in the link I gave that php is much faster and provides a script, Think I have to pay more to be able to use php on my server though so will prob go with CSV.

If you spot any nice open source terminals or anything useful open source for that matter please let me know.
The greatest force on earth is compound interest.
  • #971
  • Jun 16, 2010 7:55am Jun 16, 2010 7:55am
Quoting sebastionay
Whats an ATS?
http://en.wikipedia.org/wiki/Automated_trading_system

The old name for algo-trading
  • #972
  • Jun 16, 2010 1:25pm Jun 16, 2010 1:25pm
I started forward testing last week. Last Friday and Monday seem to be quite bad across the currencies. Has anyone that is live on the moment have the same bad results for those days?
  • #973
  • Jun 18, 2010 12:19pm Jun 18, 2010 12:19pm
What I'm currently thinking now that I've started working on the backtester...

Using a MIDI controller for rapid strategy prototyping and chart navigation.

http://glanzunion.de/wp-content/uplo...0/dsc_6857.jpg

The controls would be mapped to various settings.

For example, the last set of knobs/sliders could be used for chart navigation:
- knob for timeframe selection (from W1 -> tick)
- knob for candle width (so you can fit more on the chart)
- slider for selecting the period displayed on the chart

Another set would be for strategy control:
- slider for stop size control
- slider for profit taking control
- slider for indicator control (EMA period for example)

The moment I adjust the knobs/slider the strategy is run again for the period displayed on the chart. Candles will be colored gray for inactive periods and green/red for trades making/losing money.

In the picture is a cheap controller ($70). There are better ones which much more buttons and features.

I've done music in the past, and the productivity gains from using a hardware controller versus using the mouse to move knobs on the screen is massive. I know from personal experience how hard it is to visual backtest in MT4. What I have in mind should give significant advantages for strategy research (besides having an accurate backtester).
  • #974
  • Jun 21, 2010 12:59pm Jun 21, 2010 12:59pm
Quoting Adal
What I'm currently thinking now that I've started working on the backtester...

Using a MIDI controller for rapid strategy prototyping and chart navigation.

http://glanzunion.de/wp-content/uplo...0/dsc_6857.jpg

The controls would be mapped to various settings.

For example, the last set of knobs/sliders could be used for chart navigation:
- knob for timeframe selection (from W1 -> tick)
- knob for candle width (so you can fit more on the chart)
- slider for selecting the period displayed on the chart

Another set would be for...


Sounds like you know a lot about computers Adal, Does it take much knowledge to put together your own computer? Do you know any good websites to look at for components?

My laptop is struggling with running multiple applications. I had to order an external hard drive recently for the dukas data as its bigger than my hard drive.
The greatest force on earth is compound interest.
  • #975
  • Edited 2:49pm Jun 21, 2010 2:15pm | Edited 2:49pm
Quoting sebastionay
Sounds like you know a lot about computers Adal, Does it take much knowledge to put together your own computer? Do you know any good websites to look at for components?
LOL. I do build my own computers. Here's something to get you started:

http://www.tomshardware.com/reviews/...ng-pc,511.html
http://www.tomshardware.com/reviews/...-gpu,2656.html

I would recommend you start with one of the three configurations listed above and then customize based on your needs. The configurations are gaming oriented, so you should start with replacing the video card with a cheaper one (unless you want to do some CUDA or OpenCL) and use the money for something else - CPU, RAM or more monitors

I read reviews online to decide on the components then search around for the best price. I buy motherboard + memory + CPU from the same supplier, in case it doesn't work, since it would be difficult to figure out which of the 3 components is broken.

Advice: be VERY careful not to mismatch components, for example DDR2 memory for a DDR3 motherboard, or a Intel 1366 pin CPU for a 1156 pin motherboard. Some online advice is debatable, for example a lot of people recommend huge power-supplies (750W). I have a 450W one and it drives a quad and a hungry graphic card with no issues at all.

All the Dukascopy .BIN files (398,539 tick files at the moment) take 8.8 GB on my computer. Did you download the .CSV version?

EDIT: I remembered that I have pictures from when I assembled my current computer two years ago. It was an upgrade, I replaced only the motherboard + CPU + memory from a single core AMD to an Intel quad - http://www.dropbox.com/gallery/19021...puter?h=388914
  • #976
  • Jun 22, 2010 6:49am Jun 22, 2010 6:49am
I have an interesting question:

Is anyone trading non-profitable systems to get lower combined profit curve volatility?

I have gotten some quite interesting results by also including non-profitable systems to the possible strategy mix.
  • #977
  • Jun 22, 2010 7:41am Jun 22, 2010 7:41am
Quoting mikkom
I have an interesting question:

Is anyone trading non-profitable systems to get lower combined profit curve volatility?

I have gotten some quite interesting results by also including non-profitable systems to the possible strategy mix.
In a few interviews with automated trading guys they said that they quickly cull non-performing algorithms. I don't believe it's disinformation. I'm sure they are aware of Parrondo's paradox. That being said, the paradox is real. If my research would show that it's better overall to include a losing strategy I would do it. Science all the way. I don't expect that to happen through.
  • #978
  • Jun 22, 2010 8:21am Jun 22, 2010 8:21am
Quoting Adal
In a few interviews with automated trading guys they said that they quickly cull non-performing algorithms. I don't believe it's disinformation. I'm sure they are aware of Parrondo's paradox. That being said, the paradox is real. If my research would show that it's better overall to include a losing strategy I would do it. Science all the way. I don't expect that to happen through.
This is not that much about parrondo, much more about volatility of the profit curve.

Another way to look at this is to ask when is a combination of systems a and b a system in itself (combination of a and b might have a better sharpe than system a even if b in itself is not profitable system) and then to multiply this question to N systems.

I look this totally from view of my fitness algo, before I scapped all systems when they were "not profitable" as defined by my algo but now I also test non-profitable algos against a combined curve and sometimes there are non-profitable algos that give me better fitness than I would get without them. This is of course totally dependant on how one measures fitness.
  • #979
  • Jun 23, 2010 5:37pm Jun 23, 2010 5:37pm
A new gaming method, overload the competition (http://www.nanex.net/20100506/FlashC...s_Part4-1.html):

What benefit could there be to whomever is generating these extremely high quote rates? After thoughtful analysis, we can only think of one. Competition between HFT systems today has reached the point where microseconds matter. Any edge one has to process information faster than a competitor makes all the difference in this game. If you could generate a large number of quotes that your competitors have to process, but you can ignore since you generated them, you gain valuable processing time.

Full report: http://www.nanex.net/20100506/FlashCrashAnalysis_Intro.html
  • #980
  • Jun 24, 2010 4:47am Jun 24, 2010 4:47am
Quoting Adal
A new gaming method, overload the competition (http://www.nanex.net/20100506/FlashC...s_Part4-1.html):
Regulators seriously needs to get up to speed in understanding and banning these manipulative behaviours before the morons how utilize this destroys it for all the others who play by the book.

Sooner or later exchanges needs to rethink their intire setup in order to create a fair-play game in the age of computerized trading.
0 traders viewing now