Volume Splitter

[FulcrumTrader 10]

This kind of error could cause a trader to enter or exit based on incorrect information and just one mistake could cost more than the subscription to Market Delta & IQFeed!

 

Absolutely correct, so trying to make broker supplied feeds work for BID/ASK differential duties is an exercise in futility. Use a proper data feed (with a ticker plant and available historical BID/ASK data) for a few bucks more a day and then focus your efforts on your trading set ups.

[FulcrumTrader 10]

Here is a ladder comparison between Ninjatrader with GOM Ladder & Zen-Fire & Market Delta with IQFeed. Look at the circled area. IQFeed & Ninja have totally different results. Interestingly enough, the price above and below they are almost the same.

 

This kind of error could cause a trader to enter or exit based on incorrect information and just one mistake could cost more than the subscription to Market Delta & IQFeed!

The exceptionally well built GomLadder (another excellent job by supper Gomi!) in NT connected to DTN.IQ feed will give you results that you can at least trust.

 

BTW, did I mention that Gomi rocks!

[BlowFish 10]

Yes....a $2,000 a month feed.

 

DTN.IQ feed passes the verification test on a week by week basis......none of the broker supplied feeds do from all the various testing done.

 

I dunno this still sounds rather blanket : I am not after a detailed test plan just a couple of sentences on how you test and with what.

 

Which feed? I am not sure why the secrecy tbh Is it NxCore? Thompson/Reuter? Do you check that the cumulative total matches day to day? That would seem to be a fairly good 'checksum'. Do they tally exactly? Does it vary day to day? Are you using the same client software to test? Just a couple of sentences would elevate this from being a blanket statement.

 

Edit I am not disagreeing with your conclusions (thoughwith broker supplied feeds infrastructure plays a big part) just wondered how you arrived at them

Edited March 31, 2010 by BlowFish

[AgeKay 10]

Now, in the data feed, that low level of trade stuff does not need to be reflected, but... It seems quite reasonable to show book updates prior to transactions. It could go either way--so long as you are consistent and always do it one way or the other.

 

I would be interested... in the low level data of dtn.iq--are book updates always given before transactions? So a trade of 10 at bid will be always immediately preceded by a reduction in the bid size by 10?

 

I agree. I saw the same thing with the TT feed. Book updates always came before trades. I assume that this is also the case with IQFeed because if trades came before book updates then you wouldn't have the issue with trades between the inside market.

 

I have looked at event data for zen-fire and can say that there are definitely things that don't add up. Something like bid is at 19, there's a trade at same bid price of 25 and next update is bid is at 43 at same price. Obviously, there should have been some updates in the middle in order for that to have happened but they didn't show up in the feed I received.

 

I saw the same thing with TT's feed. Most likely they just drop the book updates in favour of showing you the most recent inside market. I've talked about his earlier. The feed providers have to decide whether speed or complete data is more important and in most cases it's speed. DTN claim that they don't filter anything. Even their API documentation says that you they can't guarantee that all the data you get is valid because they don't filter anything.

[BlowFish 10]

I always thought filtering was a cop out, a throw back to a bygone era. It is using a band aid rather than curing the wound. Put another way with robust error checking filtering should be unnecessary. I have also wondered how an electronic exchange can be designed that can generate 'bad ticks'. Sure some fat fingered human might accidentally sell 100,000 rather than 100 but that's a different matter altogether.

 

I guess when people talk of filtering they might mean coalescing or aggregation? Filtering to me implies purposely removing data.

[FulcrumTrader 10]

I dunno this still sounds rather blanket : I am not after a detailed test plan just a couple of sentences on how you test and with what.

 

Which feed? I am not sure why the secrecy tbh Is it NxCore? Thompson/Reuter? Do you check that the cumulative total matches day to day? That would seem to be a fairly good 'checksum'. Do they tally exactly? Does it vary day to day? Are you using the same client software to test? Just a couple of sentences would elevate this from being a blanket statement.

 

Edit I am not disagreeing with your conclusions (though with broker supplied feeds infrastructure plays a big part) just wondered how you arrived at them

Yes on a day to day basis the Cumulative Delta plots/data match up exactly when the feeds are compared together....we are not finding any data discrepancies.

[MidKnight 12]

You raise a good point about bid/ask in that is it really worth it to have bid/ask instead of the more reliable uptickd/downtick?

 

I'm pretty sure I remember reading at the Neoticker forum that Lawrence Chan (long time system developer, trader, and owner/creator of TickQuest, the makers of Neoticker -- What a mouth full that was!) always recommends using up/dn tick on any data feed due to its higher reliability across data sources.

 

With kind regards,

MK

[FulcrumTrader 10]

I'm pretty sure I remember reading at the Neoticker forum that Lawrence Chan (long time system developer, trader, and owner/creator of TickQuest, the makers of Neoticker -- What a mouth full that was!) always recommends using up/dn tick on any data feed due to its higher reliability across data sources.

 

With kind regards,

MK

Yes, back in the day it was not easy to get clean BID/ASK differential data (or charting software that even knew how to show that data.....INCLUDING NeoTicker). Also, after all these years you can see not even Neoticker has any decent off the shelf Cumulative Delta tracking tools. LC does not at all seem interested in CD, so I am never surprised when he makes comments like that.

 

Well now we have MarketDelta, Investor RT, NT7, Sierra Charts, and soon TradeVec and the next Tradestation summer release that will have Cumulative Delta tools.....COOL!

 

Traders who want to pay attention to the order flow are asking for these bid/ask differential tools everywhere, so I wonder how good LC is at spotting a trend?

[heretodaygone... 10]

Yes, back in the day it was not easy to get clean BID/ASK differential data (or charting software that even knew how to show that data.....INCLUDING NeoTicker). Also, after all these years you can see not even Neoticker has any decent off the shelf Cumulative Delta tracking tools. LC does not at all seem interested in CD, so I am never surprised when he makes comments like that.

 

Well now we have MarketDelta, Investor RT, NT7, Sierra Charts, and soon TradeVec and the next Tradestation summer release that will have Cumulative Delta tools.....COOL!

 

Traders who want to pay attention to the order flow are asking for these bid/ask differential tools everywhere, so I wonder how good LC is at spotting a trend?

 

That's pretty harsh coming from someone who has been calling the top since last April.(ES - 879, wasn't it?)

 

Prolly going to be right this April. You should start a thread. lol

[FulcrumTrader 10]

That's pretty harsh coming from someone who has been calling the top since last April.(ES - 879, wasn't it?)

 

Prolly going to be right this April. You should start a thread. lol

I never called any top last April....you seem confused. I have traded 9 longer term ES trades since this time last year and all were profitable (some LONG and some SHORT). Had some very nice longer term LONG trades this past year too.

 

[ame=http://www.youtube.com/watch?v=yQpQ6OZCyZA]YouTube - Delta Zone for the BUY![/ame]

 

 

 

Three very good longer term LONG trades in the ES this past year.....LONG off the 866/867's.....off the 977/978's....and was just recently in a longer term LONG ES trade off the 1040.75's bounce. I trade BOTH directions as I see longer term set ups form.

[FulcrumTrader 10]

[ame=http://www.youtube.com/watch?v=h83G32E8W-4&feature=PlayList&p=0AC673004700610C&playnext_from=PL&playnext=1&index=3]YouTube - S&P500 Emini rally off the 865's explained[/ame]

[FulcrumTrader 10]

That's pretty harsh coming from someone who has been calling the top since last April.(ES - 879, wasn't it?)

 

Prolly going to be right this April. You should start a thread. lol

Being harsh is very different from stimulating some competition in the vendor charting industry.

[MidKnight 12]

Yes, back in the day it was not easy to get clean BID/ASK differential data (or charting software that even knew how to show that data.....INCLUDING NeoTicker). Also, after all these years you can see not even Neoticker has any decent off the shelf Cumulative Delta tracking tools. LC does not at all seem interested in CD, so I am never surprised when he makes comments like that.

 

Well now we have MarketDelta, Investor RT, NT7, Sierra Charts, and soon TradeVec and the next Tradestation summer release that will have Cumulative Delta tools.....COOL!

 

Traders who want to pay attention to the order flow are asking for these bid/ask differential tools everywhere, so I wonder how good LC is at spotting a trend?

 

Hiya FT,

 

I can't entirely speak for how LC trades, but I know he has done extensive work on market internals, and that is why neobreadth exists. He is a trader, and what you see in Neoticker is what tools he and the TickQuest traders need - not necessarily what others want. The model is more about giving you the tool to make anything you want in any language you want.

 

Delta tools are available from the forum or one can code it up however they like in whatever language they like.

 

Yes, I am a Neoticker fan

 

With kind regards,

MK

[FulcrumTrader 10]

Hiya FT,

 

I can't entirely speak for how LC trades, but I know he has done extensive work on market internals, and that is why neobreadth exists. He is a trader, and what you see in Neoticker is what tools he and the TickQuest traders need - not necessarily what others want. The model is more about giving you the tool to make anything you want in any language you want.

 

Delta tools are available from the forum or one can code it up however they like in whatever language they like.

 

Yes, I am a Neoticker fan

 

With kind regards,

MK

Right, I am familiar with Neoticker (used it for a while in the past) and I do like the charting application overall. I have also followed the CD tools discussions at the NT forum in the past........... CVD (Cumulative Volume Delta) Indicator - Page 3 - NeoTicker Forums

 

I have several trader friends who use NT so I know you get some workable flexibility in the use of their charting app.

[AgeKay 10]

With regards to data feeds. I talked to the API guy of IQFeed yesterday and he couldn't tell me whether inside market updates come before trades or not. He said it depends on how the exchanges report it because they just forward it. I just called Eurex to find out and they told me they'll get back to me with an answer on how they report trades to the feed providers.

 

I compared IQFeed with TT's feed today and IQFeed serves more trades. There was even a point where MD Trader showed a low of the day without any volume having traded there, which means that they somehow filtered out that trade.

 

I've said before that ideally you want two different feeds. One that reports all information and is possibly a bit slower (used for charting and data processing) and one that reports only the most recent information and is really fast (used for execution/DOM). I do have the impression that IQFeed is the former and TT's feed is the later.

[cunparis 10]

I've said before that ideally you want two different feeds. One that reports all information and is possibly a bit slower (used for charting and data processing) and one that reports only the most recent information and is really fast (used for execution/DOM). I do have the impression that IQFeed is the former and TT's feed is the later.

 

This makes a lot of sense. I am using IQFeed for Market Delta charting and Zen-fire for Ninjatrader for the DOM.

 

I don't know if it's the Market Delta software or the IQFeed but when I compare the ninja DOM to the Market Delta Time & Sales and charts, the Ninja DOM moves faster. There is maybe 1/4 to 1/2 second lag in Market Delta. This is why I think it's important to keep an eye on the DOM.

[BlowFish 10]

Yes on a day to day basis the Cumulative Delta plots/data match up exactly when the feeds are compared together....we are not finding any data discrepancies.

 

Thanks, matching exactly is pretty conclusive and that alone would add weight to the whole procedure.

[BlowFish 10]

I'm pretty sure I remember reading at the Neoticker forum that Lawrence Chan (long time system developer, trader, and owner/creator of TickQuest, the makers of Neoticker -- What a mouth full that was!) always recommends using up/dn tick on any data feed due to its higher reliability across data sources.

 

With kind regards,

MK

 

As an aside Neoticker (in theory, I have not used NT for quite a while) would be a good option for this type of work (with its 'tick precise' technology).

[maxima 10]

I'm pretty sure I remember reading at the Neoticker forum that Lawrence Chan always recommends using up/dn tick on any data feed due to its higher reliability across data sources.

 

With kind regards,

MK

Try and plot CVD as Fulcrum does alongside with cummulative up/down tick volume delta... and look at it - can you figure out anything from it? Tell us if you could.

[MidKnight 12]

Try and plot CVD as Fulcrum does alongside with cummulative up/down tick volume delta... and look at it - can you figure out anything from it? Tell us if you could.

 

Hi there,

 

I have indeed done this for quite some time on Asian futures. I'm not sure I understand your question though. Figure anything out?

 

With kind regards,

MK

[maxima 10]

OK. I thought you havent done... because cum delta up/dn ticks makes no sense whatsoever. while cvd on bid/ask volume makes sense most of the time...

 

i tried it for iqfeed and barchart.... perhaps your experience varies... good for you!

[BlowFish 10]

Both are methods to try to detect 'trade direction'. Lots of studies have been done on how efficacious these methods are (I mentioned these earlier in the thread). Broadly speaking most conclude that V@B/A is about 10% more efficacious than uptick/downtick though results do vary. None of it is precise however they did convince me that it is a 'good enough' metric to be useful.

[maxima 10]

from my personal use of CVD - B/A has 25% efficience (which is high) and up/dn tick = 00000000... which for me doesnt mean up/dn tick 25% less efficient for me that mean - useless

 

P.S. I use very few indicators (2-3 at a time). so may be I am not the one to argue on this :helloooo:

[BlowFish 10]

from my personal use of CVD - B/A has 25% efficience (which is high) and up/dn tick = 00000000... which for me doesnt mean up/dn tick 25% less efficient for me that mean - useless

 

P.S. I use very few indicators (2-3 at a time). so may be I am not the one to argue on this :helloooo:

 

That's odd most studies have 60%-65% (ish) for up/down and 70-80% (maybe a wee bit more) for V@B/A. Of course whether trade direction is a precursor to directional change in price is another matter. This is in markets where there is an audited database of trades (e.g. specialist markets) so it is possible that they behave a little differently. I wonder whether it is your testing methodology? Of the dozen or so papers that I have found or purchased I cant recall it ever been tested below 60%. Most of the studies have very rigorous testing methodologies. I am trying to imagine where the discrepancy could come form? What data are you using? How are you determining the actual trade direction of each trade?

 

Edit: A thought occurs to me, maybe you are talking about efficacy of using Lee & Ready to determine short term price direction? The studies are concerned with using the algorithm to determine the trade direction (order flow if you like) on each individual trade. Whether this can be used to aid trading decisions is a different matter altogther.

[BlowFish 10]

I should add that my view was pretty similar to yours (Maxima) not so long ago