HI all - I have been researching some off-the-shelf solutions for equity factor attribution and exposure models and have concluded that many of the providers (namely, BBG, MSCI, Omega Point, etc) are a bit pricier than what we had expected. My firm is just introducing this type of analysis and doesn't really have the appetite to spend so much. Additionally, many of them incorporate fancy UI/portfolio mgmt software into their offerings that we really don't have any use for.
I was wondering if it is a mistake to build something internally instead. Does anyone have experience or literature that would help with this? I was thinking of using publicly published factor indices and running a lasso to determine which factors are pertinent to our securities and then an ols from there. Running fama-french 3 or 5 factor models is straightforward in python but some of these providers offer 20+ factors in their models. I wonder if I am oversimplifying this as they likely have an army of phds and i am just here by myself!
I was wondering if it is a mistake to build something internally instead. Does anyone have experience or literature that would help with this? I was thinking of using publicly published factor indices and running a lasso to determine which factors are pertinent to our securities and then an ols from there. Running fama-french 3 or 5 factor models is straightforward in python but some of these providers offer 20+ factors in their models. I wonder if I am oversimplifying this as they likely have an army of phds and i am just here by myself!
