Posts

Showing posts from April, 2024

Performance Attribution Cumulative Time Series

Image
Part IV: Cumulative Time Series of Attribution Effects In parts  I  and  II , we set up a test database with performance calculations, and in part III , calculated the daily Brinson-Fachler performance attribution results. Now, I will add another inline user-defined table function to calculate cumulative time series of all the contribution and attribution effects using CariƱo scaling . These can then be used alongside a chart of portfolio and benchmark cumulative returns to explain the difference between them. And, of course, taking just the last item of the cumulative time series gives you the full period results only. The function is quite long, some 250 lines of code, so the full version will only be available in my  GitHub repository . However, much of the length is just simple repetition, as the statement must use subqueries to calculate values in phases, and there are a lot of descriptive fields to "carry over" from one subquery to the next. Scaling required T...