Was trying to run some return analysis on my own factors for the first time, and noticed some trouble with the cumulative returns data during certain periods. These persisted when cloning and running the lecture code as well. When running during the preset range (2014-1-1 through 2015-1-1) and most other dates the graphs looked fine. However, there are certain short periods of time where the graphs spike, which throws off some of the rolling statistics I am trying to calculate for my own analysis.
Does anyone know what causes this problem, and if there is any work around? I have attached a lecture clone for reference. Only the pipeline date range is changed.