Systematic trading | Page 36 | 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

  • #701
  • Apr 9, 2010 5:25am Apr 9, 2010 5:25am
Spontaneous Synchronization

http://ohhaveyouseenthis.blogspot.co...onization.html

Just reading through a thread on ET about this.
Economists have forecast 9 out of the last 5 recessions
  • #702
  • Apr 12, 2010 2:51am Apr 12, 2010 2:51am
Quoting Craig
I don't think there is anyway around the 10k minimum, in the larger picture you're going to need a lot more than that to trade forex at IB within tolerable risk limits and reasonable commissions. Once you have the account, you can open a 'paper trade' account against which one can develop, but you still need the 10k in the 'real' account.

Sorry to pester you but I have a few questions.

1) Do you think its better to forget trading forex, given its very efficent (well the pairs with resonable spreads), and go directly to stocks? I dont ant to waste my time mining Forex data if its much easier to find somethinf that works in stocks. But I suppose starting capital may be a significant factor.

2) What would you say is min starting balance for equities (I dont mean brokers min, or PDT min, what you think)?

Regards

Mark
Economists have forecast 9 out of the last 5 recessions
  • #703
  • Apr 12, 2010 2:59am Apr 12, 2010 2:59am
In case any of you doesn't already know it, this is a really cool blog:

http://autospreader.wordpress.com/
  • #704
  • Apr 12, 2010 3:15am Apr 12, 2010 3:15am
Quoting BlackMage
In case any of you doesn't already know it, this is a really cool blog:

http://autospreader.wordpress.com/
He suggests using Python for quant analysis, hmmm.
Economists have forecast 9 out of the last 5 recessions
  • #705
  • Apr 12, 2010 4:14am Apr 12, 2010 4:14am
Quoting BlackMage
In case any of you doesn't already know it, this is a really cool blog:

http://autospreader.wordpress.com/
Excellent stuff, thanks.
  • #706
  • Apr 12, 2010 10:57am Apr 12, 2010 10:57am
Quoting markmm
Sorry to pester you but I have a few questions.

1) Do you think its better to forget trading forex, given its very efficent (well the pairs with resonable spreads), and go directly to stocks? I dont ant to waste my time mining Forex data if its much easier to find somethinf that works in stocks. But I suppose starting capital may be a significant factor.

2) What would you say is min starting balance for equities (I dont mean brokers min, or PDT min, what you think)?

Regards

Mark
1) For Pairs Trading and other 'Market Neutral' strategies, equities present many more opportunities, for GA based stuff, the jury is still out, I am in the process of forward testing the same approach that was I using on Forex. Certainly equities display the same spectrum of possibilities, in terms of some markets seem easy to optimize and others much harder (as previously discussed), the difference being is that 20 markets out of 100 that display this 'optimization' property as opposed to just 1 (EUR/USD) out of a handful. How this translates to real profit is a question which should be answered over the coming weeks.

2) US $25,000 minimum to day trade stocks (this is US law).
http://www.interactivebrokers.com/en...umDeposits.php
The breaking of a wave cannot explain the whole sea.
  • #707
  • Apr 12, 2010 11:10am Apr 12, 2010 11:10am
Quoting Craig
1) For Pairs Trading and other 'Market Neutral' strategies, equities present many more opportunities, for GA based stuff, the jury is still out, I am in the process of forward testing the same approach that was I using on Forex. Certainly equities display the same spectrum of possibilities, in terms of some markets seem easy to optimize and others much harder (as previously discussed), the difference being is that 20 markets out of 100 that display this 'optimization' property as opposed to just 1 (EUR/USD) out of a handful. How this translates...
Thanks Craig,

I think ill start with forex and try to build up to stocks, just didn't want to waste time if it was impossible to profit due to inefficiency. Got my MB demo info so its time to code.

Mark
Economists have forecast 9 out of the last 5 recessions
  • #708
  • Apr 13, 2010 6:39pm Apr 13, 2010 6:39pm
I just joined FF and wanted to introduce myself to mikkom and other systematic traders. I've been trading since 2002, initially discretionary systems, but increasingly systematic. I have a few profitable strategies in equities and bonds, now looking to diversify into forex (to add some uncorrelated return streams). I've built a Java-based backtesting and EA platform from scratch but also use off the shelf tools like Tradestation for quick and dirty backtests. Still catching up on this thread, but right off the bat, noticed that I made some different decisions, e.g., to store my security data (I use various data stores and timeframes, mostly equity, going back to < 1900). I decided on google protocol buffers for a lot of the binary storage (instead of say, mysql, which I only use for security metadata and less data-intensive structures). Then I just load the protobuf files into RAM in a distributed environment when I do backtests. Anyway, looking forward to contributing...
  • #709
  • Apr 14, 2010 12:10am Apr 14, 2010 12:10am
Hello and welcome to the party

What distributed environment are you using? Hadoop? Your own?

Any reason for protobuf? Have you investigated Thrift and Avro?

How much data do you have?
  • #710
  • Apr 14, 2010 3:01am Apr 14, 2010 3:01am
Quoting charley
e.g., to store my security data (I use various data stores and timeframes, mostly equity, going back to < 1900). I decided on google protocol buffers for a lot of the binary storage (instead of say, mysql, which I only use for security metadata and less data-intensive structures). Then I just load the protobuf files into RAM in a distributed environment when I do backtests. Anyway, looking forward to contributing...
Interesting choice, haven't seen that myself.

Just to clarify, I use mysql just as a backup storage and for more complex queries if I need them, I have all my security data cached as simple binary files.

Where do you get your data btw? tradestation?
  • #711
  • Apr 14, 2010 3:11am Apr 14, 2010 3:11am
Quoting charley
I just joined FF and wanted to introduce myself to mikkom and other systematic traders. I've been trading since 2002, initially discretionary systems, but increasingly systematic. I have a few profitable strategies in equities and bonds, now looking to diversify into forex (to add some uncorrelated return streams). I've built a Java-based backtesting and EA platform from scratch but also use off the shelf tools like Tradestation for quick and dirty backtests. Still catching up on this thread, but right off the bat, noticed that I made some different...
Welcome, I started using Java but found it extremely slow compared to C++, so I have I am doing everything in C++.

Do you find having data going back so far useful? I am only really looking at the last few months, given the market changes often.

Mark
Economists have forecast 9 out of the last 5 recessions
  • #712
  • Apr 14, 2010 3:26am Apr 14, 2010 3:26am
Quoting markmm
Welcome, I started using Java but found it extremely slow compared to C++, so I have I am doing everything in C++.

Do you find having data going back so far useful? I am only really looking at the last few months, given the market changes often.

Mark
I would say it depends on what timeframe you work. In other words the number of transaction before you have some stats that are meaningfull
  • #713
  • Apr 14, 2010 3:41am Apr 14, 2010 3:41am
Quoting spekkiefx
I would say it depends on what timeframe you work. In other words the number of transaction before you have some stats that are meaningfull
True, I am looking at intrada, tick stream based.
Economists have forecast 9 out of the last 5 recessions
  • #714
  • Apr 14, 2010 4:46am Apr 14, 2010 4:46am
Quoting markmm
Do you find having data going back so far useful? I am only really looking at the last few months, given the market changes often.
Craig has mentioned interesting stuff on this in this thread (tscore). I raised similar question like yourself and he was kind enough to share some knowledge.

It appears from my stats that the lower the time frame the lower the profit factor becomes. I guess this is quite logic as one is not waiting for more unique situations.
  • #715
  • Apr 14, 2010 5:20am Apr 14, 2010 5:20am
Quoting spekkiefx
Craig has mentioned interesting stuff on this in this thread (tscore). I raised similar question like yourself and he was kind enough to share some knowledge.

It appears from my stats that the lower the time frame the lower the profit factor becomes. I guess this is quite logic as one is not waiting for more unique situations.
It all depends on the system, profit factor of a crap system will be low on daily charts as well as 15 min charts. I use the tick data to organise it into different time groupings.
Economists have forecast 9 out of the last 5 recessions
  • #716
  • Apr 14, 2010 5:31am Apr 14, 2010 5:31am
I use forex tickdata store in my mysql and transform to 4h data in matlab..
  • #717
  • Edited 6:27am Apr 14, 2010 6:16am | Edited 6:27am
Quoting markmm
It all depends on the system, profit factor of a crap system will be low on daily charts as well as 15 min charts. I use the tick data to organise it into different time groupings.
The best I have come up with during my testing on the lower timeframe (using tick data) is a profit factor of 2 (196 transactions over 3 months). This is an optimised strategy value so I would say that 1.4 is probably the most I can achieve with this strategy in real trading.
  • #718
  • Apr 14, 2010 6:24am Apr 14, 2010 6:24am
Quoting spekkiefx
The best I have come up with during my testing on the lower timeframe (using tick data) is a profit factor of 2 (196 transactions over 3 months). This is an optimised strategy value so I would say that 1.4 is probably the most one can achieve in real trading.
Im not sure how you equate that because you could get 2 or 1.4 then thats the most anyone else on the planet could get? BTW, Im not saying I have better results as I am concentrating on infrastrucure of my ATS at the moment. I just find it difficult to follow you extraploation.
Economists have forecast 9 out of the last 5 recessions
  • #719
  • Apr 14, 2010 6:27am Apr 14, 2010 6:27am
Quoting markmm
Im not sure how you equate that because you could get 2 or 1.4 then thats the most anyone else on the planet could get? BTW, Im not saying I have better results as I am concentrating on infrastrucure of my ATS at the moment. I just find it difficult to follow you extraploation.

Fair point, I corrected the text.
  • #720
  • Apr 14, 2010 6:35am Apr 14, 2010 6:35am
Quoting spekkiefx
Fair point, I corrected the text.
No problem spekkie, just think its good to keep an open mind. Good to have this thread although its kinda slow going.
Economists have forecast 9 out of the last 5 recessions
1 trader viewing now