Peter Davies is the DOM guy. Was a frequent t2w poster. Knowledgeable guy as far as evil vendors go!! 
Futures Trading Analysis
- Thread starter Futures Track
- Start date
While there may be some general observations I would be able to state, my experience at ET has made me reticent to state anything more than the most general statements about trading in general. So there is a gap between willing and able for me at this point. Besides, there are other sources for this information much better than I.
game, here's a thread with some good information on market depth. It assumes spoofing on both sides. The thread starts getting silly after awhile.
http://www.elitetrader.com/et/index.php?threads/dom-acv-and-bid-ask-wall-used-tgether.103293/
Depth isn't the only story though. It's how traders react to the depth that is presented to them. Search cumlative delta volume.
Put market depth, delta volume, and some SLA together and you will be farther along than most.
Good Luck!
Let's try this again. Has anyone here used backtesting as part of their learning about futures trading?
If so, how did you construct your continuous contract? Thanks.
Last edited:
My first googling attempt yielded this:
Continuous Futures Contracts for Backtesting Purposes ...
www.quantstart.com/.../Continuous-Futures-Contracts-for-Backtesting-Pu... Jan 29, 2014 - .
"In a previous article on QuantStart we investigated how to download free futures data from Quandl. In this article we are going to discuss the characteristics of futures contracts that present a data challenge from a backtesting point of view. In particular, the notion of the "continuous contract" and "roll returns". We will outline the main difficulties of futures and provide an implementation in Python with pandas that can partially alleviate the problems. ..."
www.quantstart.com/.../Continuous-Futures-Contracts-for-Backtesting-Pu... Jan 29, 2014 - .
"In a previous article on QuantStart we investigated how to download free futures data from Quandl. In this article we are going to discuss the characteristics of futures contracts that present a data challenge from a backtesting point of view. In particular, the notion of the "continuous contract" and "roll returns". We will outline the main difficulties of futures and provide an implementation in Python with pandas that can partially alleviate the problems. ..."
E-mini S&P 500 Futures, Continuous Contract #1 (ES1 ...
https://www.quandl.com/.../CME_ES1-E-mini-S-P-500-Futures-Continu...
https://www.quandl.com/.../CME_ES1-E-mini-S-P-500-Futures-Continu...
Attachments
So is it safe to assume that you have never used a continuous contract yourself?
I am looking for feedback from those who have backtested futures so they can answer practical questions about the actual splicing of contracts.
Whether or not the splicing of contracts matter would depend on your questions. If daytrading or trading short-term, the "splicing" would be irrelevant unless one were holding through the rollover. If trading a longer bar interval, such as hourly or daily, the splicing would also be irrelevant.
I assume your questions have something to do with coding?
Read the article he posted. The type of splicing matters.
If you're a quant. If you're a discretionary trader, it doesn't matter. That's why I asked what your questions were.