Search
× Search
Saturday, December 21, 2024

Archived Discussions

Recent member discussions

The Algorithmic Traders' Association prides itself on providing a forum for the publication and dissemination of its members' white papers, research, reflections, works in progress, and other contributions. Please Note that archive searches and some of our members' publications are reserved for members only, so please log in or sign up to gain the most from our members' contributions.

How to Resolve Futures & Currency Spread Issues ?

photo

 Brian Wong, Quantitative Researcher at Hedge Fund

 Friday, November 21, 2014

This may be one of the common issues in algo-trading which cuts lots of Sharpe ratio of your strategies when restricted to adopt market orders; certainly, you may consider pending when placing orders but the effect of pending, especially market impact, is not easy to be verified by data without live trading Whether pends or not also relies on your algo For instance, momentum strategy is triggered by rising trend and therefore, a pending order is hard be filled What about your view of relieving the pain of spreads ?


Print

4 comments on article "How to Resolve Futures & Currency Spread Issues ?"

photo

 Marko Rantala, Indicator & Strategy Developer. Futures trading. CEO & Founder seeking new partnerships ► http://tradingmaestro.com

 Sunday, November 23, 2014



Hi Brian,

I have realized with those very short time trades that no point to set limit orders as the best trades lost and total revenue worse but just to keep the eye on spread and do not trade if spread is too wide. It will serve another aspect, automatic removal of bad timing (=news and so on) as spread starts to widen couple of minutes before those as most of the robots don't trade then. (If you have forgotten to remove the bad timing otherwise).

For that I have done a average spread indicator to NT, which I use in my live robots not just to see the last spread.


photo

 Vassil Dimitrov, Trader & Strategist at Business Partners

 Tuesday, November 25, 2014



@Brian,

yes, spread (and also broker commissions) can cut a lot of trading performance.

In order to study the best solution for your system, you need to have a very reliable backtesting software and work with past live trading data.

Up on that, you may attempt some upgrade of the systems, but if you lack the correct simulating environment, your conclusions would be "polluted" anyway...


photo

 Stefan Simik, Quant / Trading Systems Developer

 Thursday, November 27, 2014



There is one simple and effective solution, with benefit of reproducible backtest results without too much dependency on how super-sophisticated isthe trade/fill simulation engine.

I would say, that min 90% of all trades will have minimum 1 tick drawdown.

This includes also fast momentum strategies, when there is need to enter trade immediately.

So you can use limit entry orders for most cases.

If the trade will have 1 tick drawdown, you are safely in position for predictable price.

If not, you can check for moving market price and if it is too far - just cancel entry order and skip the trade. Advanced option is to move entry 1-2 ticks, if the trade is not filled in very short time.

There are many benefits of such an aproach. It mitigates risk, guarantees safe fills, produces predictable results. And there is positive expectation, that you get some fills on price-touch, which can occasionally cause some additional profits comparing to the backtest results.


photo

 Marc Levine, Chief Executive Officer, Levco Capital Markets R.E. Investment, Income Producing Properties

 Saturday, November 29, 2014



Brian

Please login or register to post comments.

TRADING FUTURES AND OPTIONS INVOLVES SUBSTANTIAL RISK OF LOSS AND IS NOT SUITABLE FOR ALL INVESTORS
Terms Of UsePrivacy StatementCopyright 2018 Algorithmic Traders Association