Supporting Quantitative Research is a major use case for our Velocity product suite, so we have a great deal of contact with Quant users. Over the last two years, we have seen an explosion of R usage within this community. While we still have a large number of clients using Matlab with Velocity, and we don't see Matlab going away at all, there's no denying the rapid ascent of R.
Given the open source nature of R, and the community around it, the usage seems to be snowballing. New add on packages are being created all the time. The cost makes it extremely compelling for Universities to use on their Financial Engineering courses, and so we are now seeing people come out of higher education with extensive R experience. When these people join Wall St, whether at broker / dealers or hedge funds, their first instinct is to use the statistical analysis package they already know.
Another development is the recent emergence of commercial distributions of R, analogous to how Red Hat provide a commercial distribution of Linux. One such company, REvolution Computing, provide a commercial distribution called RPro and are also doing very interesting things with parallel processing and R.
For those not familiar with R, it's an open source implementation of the S programming language. You can download for free at the R Project page . The S language is implemented in the commercial package S-Plus from Insightful Software, who were recently purchased for a song by Tibco. It is the belief of ex-Insightful employees I have spoken to that Tibco are mostly interested in the Insightful server products - the Data Miner and S-Plus Server. As such, the future of the SPlus desktop product would appear to be in question. We expect that to further fuel the demand for, and usage of, R in Quantitative Research.
About a year ago, Vhayu made the decision to commit to R as our time series language. We felt that coming up with another proprietary language (a la FAME, LIM, Q) would be a dead end for both Vhayu and our customers.
We launched our Velocity for Quant Research package in April this year at TradeTech in Paris. This desktop package aimed at the Quant community includes the first phase of our R integration, we have committed to making our integration with R the fastest and most complete of all tick data storage vendors. The interest in, and sales of, this package have been incredible. We see this as a real validation of our approach - leveraging existing R experience with the ease and speed of our Velocity engine to help deliver Alpha faster.
Here's a simple example to merge two trade time series for different symbols and produce a derived time series where the spread between the two instruments exceeds 0.5
xy <- na.locf(merge(x, y))
xy[abs(xy$x - xy$y) > .5, "y"]
While this may not be intuitive to everyone on first reading, the important thing is that it is instantly intuitive to those members of the R community doing time series analysis. No learning curve for proprietary languages needed. Our R adapter can be downloaded from http://www.vhayu.org
Our increasing integration with R also touches on another topic, the gravity of data, which we'll cover soon.
Comments