Damn and Doh! You are exactly right. NT is the software and I get the data feed from AMP. MIRUS uses the same feed as AMP. Thank you for the correction. lj
For purposes of clarification only http://www.elitetrader.com/vb/showthread.php?threadid=139218: Apologies to all who might have read the above post. As is frequently the case, as soon as you say something hasn't happened, it happens. This from a responder on the SUTTTJHM thread and my response to him: Damn and Doh! You are exactly right. NT is the software and I get the data feed from AMP. MIRUS uses the same feed as AMP. Thank you for the correction. lj Bottom line the question remains: Can two RLC feeds from different data vendors be expected to be the same? lj lj
lj: I use TradeNavigator for my charting and NinjaTrader for my DOM and trading. The reason is that my broker, MB Trading, is a partner with NT and not TN. TN uses their broker feeds for their DOM. To answer your question regarding any difference in data feeds, I see at a volume difference with each of these. The TN volume is consistently higher than that from BM Trading (through NT). I previously used QCharts which allowed users the option to choose different servers for the data feed. Their volume differed depending on their servers.
Thanks brandx. Yeh I brainfarted and neglected to point out that it was the Zenfire datafeed that went 'through' AMP (and brokers like Mirus, MB Trading and others) to my NT software. The brokers don't mess with the data - I am almost sure of that. The way I understand it the 'raw' RLC data leaves the CME, goes to various high end data vendors and from them to the user, frequently 'through' a broker. It may be more complicated than that , and probably is, but I was curious to see whether traders had noticed any differences in the time bars and/or volumes depending on the data vendor. It appears from what you say that there is. I have spoken with CME and they state that that the RLC data is the same for everyone, so if they're being truthful, then the Zenfire folks and the Genesis crew are doing some sort of tweaking. If you wouldn't mind me asking, have you had occasions where the data differences would have affected your decision to enter/exit or reverse a trade. TIA lj
lj: If one uses volume levels rather than comparative volume bars, one could look at a dry-up situation that wasn't really dry-up. I always like a higher volume feed. I also noticed that T & S quantities are also effected. I speak with another trader as I am watching the T & S and we get different sized trades going through at the same time. I hear that bundling of orders can cause this.
Thanks again brandx. That T&S thing is, I think, bundling of 'late trades'. It does not affect the DOM though and that's important. Best of luck to you and good trading too. lj
ljyoung, there will always be differences. Some of them will be due to how your feed vendor handles incoming data and how they're connected to the exchanges, and some of them can be explained through simple stuff like how they handle timestamping (if they indeed do stamp tick data, not all do). But, FWIW, this is part of an email response from genesis back in august: (I also subscribe to historical tick data from several vendors. When they've had time to "scrub" the data, the differences can become even more disturbing. One would think it would be the other way round. ) Anyway, IMHO the differences will not seriously affect the 'price volume relationship'. Thank you .
Thanks for your considered response. As well you know the Zenfire and Genesis datafeeds are at the higher end of the spectrum of feed vendor product. The crew that peddles the mid to low end stuff should be eviscerated but that's another story. It is good to hear as well from a Hershey practitioner that the PV relationship is not seriously compromised by data corruption. As I implicitly mentioned to brandx what one doesn't need is the above variable adding to the bad trade entry point I usually decide on. Just for chitty-chatty, if the OP doesn't object, do you know at what point timestamping occurs when a retail trader has a direct (and rather spendy) connection to the exchange? Client desktop software vs 'exchange-originated'? Regards, lj
Yes, you very much get what you pay for (but sometimes not even that). No, you would have to investigate that per instrument per feed. And the feed vendors are not keen on supplying that kind of information. To complicate things further, many feed vendors source data in turn, from companies like Interactive Data (comstock). Anyway, from my experience, I know of data that has been timestamped at the exchange, at the vendors sourced tick plant, at their own tick plant and finally timestamped at your end of the line. You just have to hope all the involved server clocks are set correctly. IMHO these timestamps are important if you're researching trading systems with years of tick data. When you're trading in real time you get all the data as fast as the feed allows (better feed, less lag...) and the actual timestamp, as long as you don't have a congested feed, is not really that important. Depending on how far down the rabbit hole you are, of course. I don't look at tick charts . Tums, sorry for being a bit off topic, feel free to request this to be deleted if you find it too far off .