Hi,
I am trying to calculate the spearman correlation of each factor to Return, and I would like to calculate it per each asset.
(as opposed to calculating it against a particular asset). Like this:
r = Returns(mask=universe, window_length=PREDICTION_WINDOW+1)
f = MyFactor()
f_past = Factor_N_Days_Ago([f(mask=universe)], window_length=PREDICTION_WINDOW+1, mask=universe)
f_r_corr = f_past.spearmanr(target=r, correlation_length=MONTH)
Here MyFactor and Factor_N_Days_Ago are the factors I implemented (Factor_N_Days_Ago basically picks up the first row).
When I add f_r_corr to the pipeline, the backtester does not proceed, only to give out an "known error" message after a while.
Any idea why this is happening?
Thank you.