This presentation is aimed at all those working in commercial and government analytics, irrespective of what tools they use, and also to those students intending on such a career. R and other open source tools play a powerful, unique and disruptive role in business analytics, and are even now changing the landscape. The use of such tools leads to new business models and strategies that challenge the conventional approach.
The key element of the presentation is “Analyst First”, a new approach to analytics, where tools take a far less important place than the people who perform, manage, request or envision analytics, while analytics is a non-repetitive, exploratory and creative process where the outcome is not known at the start, and only a fraction of efforts are expected to result in success. This is in contrast with a common perception of analytics as IT and process.
During the first part of our meeting, Nicolas Christou gave an introduction of statistical finance in R, and presented a package he co-authored with previous PhD student David Diez (2010). Video of the talk is below:
During the second part, we accommodated shorter talks outlining R users’ experiences with statistical finance in R.
Kyle Matoba, a Finance PhD student from UCLA Anderson School of Management, presented on Algorithmic Trading with R.
Bryce Little, UCLA alum, presented on Constructing Minimum Variance Portfolios with R.
Many modern data analysis problems in both industry and academia involve building a model that can predict the future based on historical variables. The 2009 KDD Cup was an international data mining competition devoted to this type of problem, where contestants attempted to predict the behaviour of mobile phone customers using an extensive database of historical information. The University of Melbourne team managed to win one part of this challenge, using R almost exclusively. In this talk I’ll give some background to the area and the specific problem, and discuss how we went about building our models. The talk will be fairly accessible, and deal with many of the practical issues encountered in this type of work.
During the first part of our meeting, Ryan Rosario presented on the topic of large datasets in R. Video, slides and code of the talk “Taking R to the Limit: Large Datasets” by Ryan Rosario at the Los Angeles area R Users Group in August 2010 are below.
Slides are also available for PDF download here.
R code is available here.
More information about the talk can be found here.
During the second part, Trividesh Jena presented on creating models in R for use with the Zementis ADAPA product in the cloud. The video of his talk is below.
Videos of the invited talks of the useR! 2010 conference as follows (courtesy by Kate Mullen and NIST). This site also aims at collecting the materials (video, slides, R code) of local R users group (RUG) meetings and various other R talks and bringing them to the larger R community. See more videos here, and if you’d like to contribute with materials, see more information here.
Welcome by NIST and
Frank E. Harrell Jr: Information Allergy
Mark S. Handcock: Statistical Modeling of Networks in R and
Panel Discussion: Challenges Bringing R into Commercial Environments
Luke Tierney: Some possible directions for the R engine and
Diethelm Würtz: The Hull, the Feasible Set, and the Risk Surface: A Review of the Portfolio Modeling Infrastructure in R/Rmetrics
Friedrich Leisch: Reproducible Statistical Research in Practice and
Uwe Ligges: Prospects and Challenges for CRAN – with a glance on 64-bit Windows binaries
This is an older (2009) video from the kickoff meeting of the San Francisco Bay Area R Users Group. It was a panel discussion within the Predictive Analytics World conference. Video courtesy by Ron Fredericks of LectureMaker (click on the image below to see the video on LectureMaker’s site).