I am currently attempting to gentrify an algo that someone else wrote in 2014. As would be expected, most of the built-in functions are now deprecated and I'm having a very difficult time finding old documentation for this specific data fetch function in order to bring the algo up to date. The code is as follows:
I understand that set_universe simply used to pass a list of up to 10% of the market to 'data.' The problem is that I do not understand what 'DollarVolumeUniverse' pulls from and how it sorts it's data. With Pipeline, I am forced to use the built-in factor 'AverageDollarVolume,' which selects from your chosen percentage range, but it first finds the average of each security's dollar volume over a given window length. So my question is, what would be the best way to fit 'AverageDollarVolume' to 'DollarVolumeUniverse'? I know that the window length probably differs, because I have tried a number of common window lengths with AverageDollarVolume and I continue to get differing results.