[GSoC-PortA] Progress Update

[Ross Bennett]

All,

Here is a brief update of my progress to date.

Factor Model Moment Estimates
Estimating the covariance, coskewness, and cokurtosis matrices based on a
statistical factor model as described in K. Boudt, W. Lu, B. Peeters,
"Asset Allocation with Higher Order Moments and Factors Models" has been
implemented.
See demo/higher_moments_boudt.R

Rank Based Optimization
Portfolio Optimization based on the Almgren and Chriss paper has been
implemented. Expressing views on the relative ranking of assets using
Meucci's Fully Flexible Views framework has been implemented.
See demo/relative_ranking.R

Multilayer Optimization
Support for optimization with a main portfolio and layer of sub-portfolios
has been added.
See demo/multi_layer_optimization.R

Support Bayesian Optimization
Support for Meucci's Fully Flexible Views framework has been implemented. I
have also added support for a basic Black-Litterman model.
See demo/meucci_ffv.R

Regime Switching Models
A general framework for specifying a regime and different portfolios for
each regime has been implemented.
See demo/regime_switching.R

TODO
Add support for supervised learning solvers
Vignette on custom moments and custom objective functions
Reformat/revise demos to compile as Rstudio notebooks

Let me know if there are any questions.

Regards,
Ross
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/gsoc-porta/attachments/20140703/55d61c09/attachment.html>