I came across this algorithm write-up:
Robust Median Reversion Strategy for On-Line Portfolio Selection
On-line portfolio selection has been attracting in-
creasing interests from artificial intelligence com-
munity in recent decades. Mean reversion, as one
most frequent pattern in financial markets, plays
an important role in some state-of-the-art strate-
gies. Though successful in certain datasets, ex-
isting mean reversion strategies do not fully con-
sider noises and outliers in the data, leading to
estimation error and thus non-optimal portfolios,
which results in poor performance in practice. To
overcome the limitation, we propose to exploit the
reversion phenomenon by robust L1-median esti-
mator, and design a novel on-line portfolio selec-
tion strategy named “Robust Median Reversion”
(RMR), which makes optimal portfolios based on the improved reversion estimation. Empirical re-
sults on various real markets show that RMR can
overcome the drawbacks of existing mean reversion
algorithms and achieve significantly better results.
Finally, RMR runs in linear time, and thus is suit-
able for large-scale trading applications.
Any interest in collaborating to code it? I think I'd prefer to work on github, versus the new Quantopian tool (although we could try it), since we'd have version control, bug reporting, etc.