Open ghost opened 3 years ago
Hi devchild, no plans due to lack of time. Still, the paper shows that it performs worse than others in terms of sharpe ratio, so I don't see much point in adding it.
I guess it depends on the dataset, there's no real comparison in the paper. The algorithm seems very interesting as it exploits both reversal and momentum...Anticor only exploits reversal. I hope someone else finds the time to add it to this library...
Any plans on adding... It's a variation on Anticor...
paper can be found https://doi.org/10.1016/j.physa.2019.04.185