I am currently working with a pretty large panel dataset containing the investment holdings of many individuals over time (i.e., for each individual I know the positions per stock over time). I was wondering what the best measure is to examine portfolio allocation changes over time. In the mutual fund literature, researchers often use flows, but I don't think that is applicable in this case of individual portfolio allocation changes.
The ultimate goal is to examine how investors' portfolio allocations change throughout the COVID crisis. Does anyone have any recommendations what the best way would be to go about this? Moreover, let me know if there are academic papers analyzing individual portfolio allocation changes -- this may give me some inspiration.