your column dict in return is empty, that might be a start :-)

Hello,
can you tell me what "column dict" is, please? This is the first time I try custom factor.
Loi

Hello,
Why did you need this line below? What is the purpose of axis 0?
std = np.std(vol, axis=0)
Loi

Sorry, I didn't think that example was helpful, so I deleted the post. But, the np.std is using numpy's std function.

Well,
If I do this:

class LF_CFO_to_total_debt(CustomFactor):
inputs = [morningstar.cash_flow_statement.operating_cash_flow]
window_length=1
def compute(self, today, assets, out, LF_CFO):
out[:] = LF_CFO

or this:

class LF_CFO_to_total_debt(CustomFactor):
inputs = [morningstar.balance_sheet.total_debt]
window_length=1
def compute(self, today, assets, out, LF_total_debt):
out[:] = LF_total_debt

My result would be a list of stock symbols. However, when I do the division between those two argument, the result return [ ][ ]. I'm puzzled.

Hi Loi, following this post here: https://www.quantopian.com/posts/best-practices-for-comparing-annual-data I am getting an output of stocks. The main thing I did was remove the screen and added a column in Pipeline()

Anyway, hope this helps!

from quantopian.algorithm import attach_pipeline, pipeline_output  
from quantopian.pipeline import Pipeline, CustomFactor  
from quantopian.pipeline.data import morningstar  
from quantopian.pipeline.filters.morningstar import IsPrimaryShare  
import numpy as np

def initialize(context):  
    attach_pipeline(make_pipeline(), 'my_pipeline')

def make_pipeline():

    primary_share = IsPrimaryShare()  
    common_stock = morningstar.share_class_reference.security_type.latest.eq('ST00000001')  
    not_depositary = ~morningstar.share_class_reference.is_depositary_receipt.latest  
    not_otc = ~morningstar.share_class_reference.exchange_id.latest.startswith('OTC')  
    not_wi = ~morningstar.share_class_reference.symbol.latest.endswith('.WI')  
    not_lp_name = ~morningstar.company_reference.standard_name.latest.matches('.* L[. ]?P.?$')  
    not_lp_balance_sheet = morningstar.balance_sheet.limited_partnership.latest.isnull()  
    have_market_cap = morningstar.valuation.market_cap.latest.notnull()

    tradeable_stocks = (  
        primary_share  
        & common_stock  
        & not_depositary  
        & not_otc  
        & not_wi  
        & not_lp_name  
        & not_lp_balance_sheet  
        & have_market_cap  
    )

    factor = LF_CFO_to_total_debt(mask=tradeable_stocks)  
    factor.percentile_between(30, 100)

    return Pipeline(  
        columns={  
            'factor': factor  
        }  
    )

class LF_CFO_to_total_debt(CustomFactor):  
     inputs = [morningstar.cash_flow_statement.operating_cash_flow, morningstar.balance_sheet.total_debt]  
     window_length=1  
     def compute(self, today, assets, out, LF_CFO, LF_total_debt):  
         out[:] = LF_CFO/LF_total_debt

def before_trading_start(context, data):  
    print pipeline_output('my_pipeline')

def handle_data(context,data):  
    pass  

ah,
Thank you for your help, Zippy. I found out that I forgot to account for LF_total_debt = 0.

Thanks Loi :-)

I would like to mention that code runs extremely slow and is not for production. You need to figure out a quicker way to pull down the data you need. At least run it only one time a month.