bayesalpha
bayesalpha copied to clipboard
Bayesian models to compute performance and uncertainty of returns and alpha.
Not sure what other plots we'd like... I can add more plots to this PR if there are other figures that would be useful.
It looks like #53 introduced yet another bug. I can't really figure out what's causing it though... ``` --------------------------------------------------------------------------- TypeError Traceback (most recent call last) in () 1 # Save...
https://github.com/quantopian/bayesalpha/blob/master/bayesalpha/dists.py#L52 not sure that line does what is expected `pm.MvNormal.dist(mu=mu, chol=chol).random(point, size)` maybe?
If only a single algorithm is being used it's a bit cumbersome to create the right meta-dataframe.
Just to have some basic example usage / overview / tutorial. Best with some innocuous returns / alpha series.
Specifically to https://github.com/quantopian/bayesalpha/blob/master/bayesalpha/model.py#L291