Hi people,
Given the recent fall of trading fees at major brokers worldwide (mainly US) the following high frequency strategy could be relevant again. Neither the strategy nor the programming was done by me (I am kind of useless!). The strategy was first proposed in this paper http://dspace.mit.edu/handle/1721.1/59122, while the algorithm was first implemented and shown in this post https://www.quantopian.com/posts/intraday-high-frequency-basket-trading-strategy (deleted user). In my opinion, the current problem with the strategy is a lack of leverage; the algorithm at the moment operates with just ~0.10x leverage. I am well aware that HTF is dominated by prop shops and it's very dependent on hardware infrastructure, but it would be nice to make a viable algorithm, at least theoretically