Closed gusgordon closed 6 years ago
Is inf the desired value? If so, this seems good.
Yeah - spoke with Thomas about it and he was also OK with it being inf. The nan value is unreliable anyways due to float error, eg ep.sharpe_ratio([1, 1, 1, 1, 1])
is inf but ep.sharpe_ratio([0.1, 0.1, 0.1, 0.1, 0.1])
is a large number.
Looks like a nice addition to me, so we can know what to expect. Any thoughts @llllllllll ?