I'm trying to create a limited version of a running pipeline algo that would just be restricted to a handful of etf's. The algo I've modified from one developed by Anthony Garner (thanks Anthony) which I like a lot. But that algo addresses hundreds of stocks at a time, and consequently would only suit large portfolios. I want something that would select the best 3 out of about 12 etf's as opposed to individual stocks.
Based on comments seen on this forum, my first attempt was to use Update_universe function and limit it to my list of etf's. This produced no error but did NOT filter by the etf's I desired.
Since that failed I though I could add a column to the pipeline that stored the symbol so that I could later
filter the resulting dataframe by that set of symbols. Here's my stab at creating the factor:
class Primary_symbol(CustomFactor):
inputs = [morningstar.company_reference.primary_symbol]
window_length = 1
def compute(self, today, assets, out, primary_symbol):
out[:] = primary_symbol[-1] But this gives me the following error code:
ValueError: setting an array element with a sequence.
What am I doing wrong?
Thanks
