Search
× Search
Tuesday, November 19, 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.

How to Trade the RSI Using a Support Vector Machine

photo

 Tad Slaff, Co-founder/CEO at Inovance

 Tuesday, November 4, 2014

The Relative Strength Index, or RSI, is one of the most common technical indicators. It is used to identify oversold and overbought conditions. Traditionally, traders look for RSI values over 70 to represent overbought market conditions and under 30...


Print

5 comments on article "How to Trade the RSI Using a Support Vector Machine"

photo

 Manuel S., Investment Management at Merrill Lynch

 Wednesday, November 5, 2014



Great piece, Tad. Clear and concise.


photo

 Tad Slaff, Co-founder/CEO at Inovance

 Wednesday, November 5, 2014



You are absolutely right, Nikitas. Including more inputs and combining multiple SVMs into an ensemble is very likely to produce more robust results. I am also a big proponent of GAs; however, using both GAs and an ensemble method does increase the complexity and make the system more of a "blackbox". It is up to you to find the right balance between sophistication and comprehension.

Glad you enjoyed the post and please keep us updated if you do go through with extending the work!


photo

 Sean Yang, Quantitative Trader at Shell Energy Trading

 Thursday, November 6, 2014



You could use boost to select the SVMs to get an emsemble of strategies.


photo

 Nikitas Goumatianos, Software Engineer, Contracting/Program Manager, Algorithmic Trader/Researcher

 Friday, November 7, 2014



Yes, boost could be used per committee of SVMs. At the moment I am using simply the voting method (each vote of each SVM counts the same). At the top, there is an strategy algorithm which processes the votes of each committee of SVMs (totally 8 different for 4 different time periods ahead) and makes trading decisions.


photo

 Michael Kunstel, Global head of Equities Quant IT at UBS

 Monday, November 10, 2014



Support Vector Machines have been to some degree been superseded by the Relevance Vector Machine. Ultimately though, perhaps spend more time on getting good inputs (and cleaning those inputs) rather than worrying about magic classifier X being better than classifier Y or committee Z.

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