Search
× Search
Thursday, March 28, 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.

Trading and Big Data

photo

 Kevin J Coogan, founder at zettacap

 Wednesday, May 22, 2013

Do you use new forms of data, such as social media and web-derived data, in trading? What are your expectations going forward for big data? Who do you see big data analytics benefiting? This anonymous survey (http://www.surveymonkey.com/s/5W368HR) will form part of my presentation on the topic at the Hadoop Summit (http://hadoopsummit.org/san-jose/program/). The 9 question survey should take only a few minutes and will help to shed light on this often mentioned but little understood intersection of finance and big data. Thanks!


Print

5 comments on article "Trading and Big Data"

photo

 private private,

 Monday, June 3, 2013



That topic is widely discussed in Germany, also for ethical reasons. Big data is already a reality in finance, but not yet widespread in finance. My team tackles it from the multi-sensor data fusion side, a technology that is not commercially used, but in use by governments (military and intelligence services). There is decent technical challenges coming with big data; for example, the diversity of data, when fundamental data are being merged with technical data; data from non-commensurable data sources that have different levels of reliability, too (e.g. Reuters more reliable than Twitter), and there is also the problem of latency (some sources are faster than others, and how does an algo know that the later data was already factored into a trading decision earlier when being communicated by a different source, eventually even in different format, too). All this, together with the need to have a knowledge about the statistical significance of data (is Twitter significant for trading at all?) is a decent technology challenge. @Christopher, don't, it is not straight forward to find a working technical solution for extracting sentiment from tweets; just take your "it doesn't work" as an interim research result and don't be discouraged solving the task. - In other words: It is solvable, but you need a whole bunch of "non-standard" technology and time and money. And big data requires big money. It is not a commodity.


photo

 Kurt Myers, Chief Technology Officer

 Monday, June 3, 2013



@Christopher, I can write many trading algorithms that don't work.

Back in the day, I recall a major satellite TV provider would publish their subscriber numbers in a weekly trade journal. When the quarterly earnings reports came out, the stock would pop on the new subscriber numbers. At some point, they stopped reporting their subscriber numbers in the journal, yet it would be a relatively easy exercise to analyze Big Data sources to estimate their subscriber numbers.


photo

 Christopher Reeves, Programmer at A2X Capital

 Monday, June 3, 2013



@Kurt, I have found that twitter data is not that reliable for building sentiment because there are too few samples, and they often don't reflect price action. I don't disagree that "Big Data" is useful to stock trading, but I think that analyzing better sources of data are a more reasonable approach.

Why should I work hard to extract sentiment from twitter when I can just look at the order book and know exactly what someone thinks of a certain stock?

Also does anyone have a twitter sentiment algorithm that has made reliable profits for you?


photo

 Thanos Koukoutsakis, Sophis 100+years Experience / CrowdFunding / Biotechnology Investment Valuation / Big Data and Social Network Analytics

 Saturday, June 8, 2013



Hello everybody,


I definitely believe that trading strategies can be developed using Big Data. It is obviously not easy. We are planning to spin off a new company Big Data Shares Strategy, and raise funding at Seedrs. Our campaign kicks off on the 18th of June. If you are interested in our progress you can "Follow" the LinkedIn Company: http://www.linkedin.com/company/3181485?trk=tyah


or join the relevant LinkedIn Group:


http://www.linkedin.com/groups/Big-Data-Shares-Strategies-5056027?trk=myg_ugrp_ovr



We already got funding from the Technology Strategy Board in the UK, and having a very strong team of academics and practitioners, we will be raising Crowdfunding at Seedrs on the 18th of June.



Regards


Thanos


photo

 Igor Gonta, CEO of Market Prophit (www.marketprophit.com)

 Tuesday, June 11, 2013



I just launched my new site that data mines social media for financial-market related conversations and generates sentiment signals and buzz. Please check it out at www.marketprophit.com

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