I am looking for a data vendor, that has real time and historical futures tick data from all major exchanges. I require the easiest possible interface, customizable queries, cleaned relatively error free data, bid depth etc...to make data gathering quick and simple. Cost is no object, I seek the very best. It is for institutional purposes. Any advice would be greatly appreciated. Hope to hear from you soon.

132 views not one reply, you guys must be thinking deeply on this one, LOL. Please keep in mind that any advice would be greatly appreciated, thanks guys, hope someone rings in.

Reuters provides a consolidated tick feed from all exchanges, major and obscure - be it NYSE or Karachi. However: - The feeds are "real time", but not very low latency. Moreover, for most of European stock exchanges and even US futures exchanges all ticks are time stamped with low resolution. Not 1 millisecond, but 1 second, if not worse. So, latency does not really matter for them. Of course, for many developing countries latency does matter in principle - what milliseconds, if not all of these excahnges even trade continuously.- The Reuters data are not very clean - my guess it that Reuters just sends what it receives from sources. - Reuters does not normalize data, that makes it difficult to work with them. Particularly: -- What fields are supplied for ticks, as well as the number and format of the fields depends on country and/or on exchange-- Trade volumes are reported different way by different exchanges : usually per trade, but sometimes cumulative for the day from market open -- Some exchanges report two-side quotes (buy and sell), other one-side quotes, other a mix-- A trade often is reported on the same tick with a quote-- The data formats change time to time, per exchange-- Reuters not enforces its specifications. For example, per specification not more than N fields can be supplies for a single tick, but if one of exchanges decides to send N+1 fields, Reuters passes them to users. Depending on how user's codes were designed, it is not absolutely unusual if the codes crash or corrupt data when it happens.Even if you know where to look, sorting out the irregularities would require may be 12 man x months, hardly less than 6. However, it can be done much faster if you limit yourself to may be 20 major exchanges and do not care about Bogota and Colombo.I do not know who is selling historical ticks data for all major exchanges. NYSE is selling TAQ dataset of ticks for US cash instruments.

DOUBLE POST

Thomson Reuters sells a product which has historical tick data, it's called Data Scope Tick History if I'm not wrong. They also have different products to get real time data and feed algo trading applications. The latency and characteristics of these products changes with the price. More you pay better data/low latency you get.Hope it helps,jp

what is the price of reuters contract +/- of bloomberg?

RTRS and BBG will not be enough, I was looking for something much better. I am still open to any suggestions, thanks for weighing in guys.

For real-time data Reuters and Bloomberg are you best hope. There are other companies but they only support a subset of the exchanges you are probably interested in. Also one data vender might have better / faster prices for exchange A and B but slower then their competitors for C and D. I am assuming you only want one connection instead of several.You say Reuters and Bloomberg are not enough, how about giving a reason why they don't solve your problem. If you give more information about what exactly you need then maybe someone can help you.Ideally you want to connect to every exchange and process the prices directly but that isn't an easy thing to do and then manage.

Such a requirement is very hard to get unless you are sat within a major institution with an existing data infrastructure and a whole department dedicated to data integrity. Reuters and Bloomberg represent the gold-standard in terms of data quality across a broad scope. There are boutique vendors serving very specific niches and asset classes, but hard to point you in the right direction without knowing specifically what you need, and you face the challenge of each having their own delivery protocol and data structures. Oanda for example has excellent tick level data for foreign exchange going back a very long time. You will find that this stuff (particularly at the tick level), is astoundingly expensive.We are working with Thomson Reuters to make available a "Data-as-a-Service" capability via our website whereby you can gain access to their DataScope Select data over the web, with access on a "light" subscription basis at various price points. The whole idea is to provide a model whereby people without access to data infrastructure such as start-up funds and researchers can get access to institutional quality data from any vendor who is prepared to expose it on service-based model.We are currently in the final stages of delivering this, and we/they are looking for early adopters to help test and refine this capability, and one of the benefits is access to free data during their beta program- they term it "early access". If you would like to be considered for participation in this please drop me a line. There are various stipulation about who you are and what type of data you need access to, as the service is limited ins cope during the early phase, but the intention is to broaden it out once the rough edges have been knocked off but can talk through that. It would ideally also require access to someone with a reasonable level of technical capability at your end. The key is to represent a real-life use case from the target demographic, and to want something the service can currently deliver in terms of the scope of the data. Hope this is of help.

Thomson Reuters provides a full tick history offering global cross asset coverage going back to Jan 1996. This service provides all fields available from the exchange integrated with full corporate actions and reference data. It provides 2 time stamps - 1 from the exchange (whatever exchange resolution) and its own millisecond timestamp back to 1996. This data is normalized across all markets and asset classes. This data is superior in quality to exchange data like NYSE TAQ from a resolution perspective as well as other providers like tickdata.com which arbitrarily rolls data into random bars!

QuoteOriginally posted by: yurakmReuters provides a consolidated tick feed from all exchanges, major and obscure - be it NYSE or Karachi. However: - The feeds are "real time", but not very low latency. Moreover, for most of European stock exchanges and even US futures exchanges all ticks are time stamped with low resolution. Not 1 millisecond, but 1 second, if not worse. So, latency does not really matter for them. Of course, for many developing countries latency does matter in principle - what milliseconds, if not all of these excahnges even trade continuously.- The Reuters data are not very clean - my guess it that Reuters just sends what it receives from sources. - Reuters does not normalize data, that makes it difficult to work with them. Particularly: -- What fields are supplied for ticks, as well as the number and format of the fields depends on country and/or on exchange-- Trade volumes are reported different way by different exchanges : usually per trade, but sometimes cumulative for the day from market open -- Some exchanges report two-side quotes (buy and sell), other one-side quotes, other a mix-- A trade often is reported on the same tick with a quote-- The data formats change time to time, per exchange-- Reuters not enforces its specifications. For example, per specification not more than N fields can be supplies for a single tick, but if one of exchanges decides to send N+1 fields, Reuters passes them to users. Depending on how user's codes were designed, it is not absolutely unusual if the codes crash or corrupt data when it happens.Even if you know where to look, sorting out the irregularities would require may be 12 man x months, hardly less than 6. However, it can be done much faster if you limit yourself to may be 20 major exchanges and do not care about Bogota and Colombo.I do not know who is selling historical ticks data for all major exchanges. NYSE is selling TAQ dataset of ticks for US cash instruments.Any more lights on the quality of Reuters tick/HF data?

QuoteOriginally posted by: takingalphaRTRS and BBG will not be enough, I was looking for something much better. I am still open to any suggestions, thanks for weighing in guys.You are incorrect. Bloomberg's B-Pipe is an excellent solution for what you are inquiring about and will be plenty. Bloomberg will size up the bandwidth depending on the number of users you have so there should be no bandwidth issues. You really can't get perfectly "clean" data feeds - there will always be prints outside the market that filter in but you can just parse the data and tell your code to exclude the outside prints (they are flagged differently). What is your application? I use Bloomberg and just went through a demo of Thomson One - turns out that Bloomberg is slightly cheaper for the individual user (1-5 users) but then Thomson gets to be cheaper than Bloomberg when you get over 5 users. I did not actually get to work with Thomson's raw data feed (their equal product to B-Pipe) but it is supposedly pretty good. If you need low latency feeds look at Activ, though Activ does not allow for historical queries, one solution might be to record going forward and build a database. Thomson's data is in a slightly different format from Bloomberg so choose one and go with it. I'm curious why you think you need something "faster or better" than a B-Pipe or Activ feed??? At that level you could subscribe to an exchange (become a member) and subscribe to their order books - or just look into colocation to cut down the latency issues - regardless you really can't get much faster than the Thomson/Bloomberg/Active feeds.