Performance Pernalty for using very large Exclude Watchlists?

AFL Programming
1

Is there a significant performance penalty for using very large Watchlists as Exclude Watchlists?

For example, I have a database with all de-listed and active symbols for the U.S. market for several decades, so the total number of symbols excludes 20,000. However, I often wish to backtest for a shorter period than the database supports, say 10 years. I can easily create an Exclude watchlist that contains all symbols de-listed prior to the start of my “shorter term” analysis, but the watchlist could exceed 10,000 symbols.

Is using the large Exclude Watchlist the best way to limit the data access required to run the shorter term analysis, or should I create a target watchlist containing only the symbols relevant to the shorter time frame?

Thanks,
David W

2

It does not matter if you use Include or Exclude filter, the performance in Analysis window is pretty much the same.

3

Thanks, Tomasz.

I had run a couple of benchmarks and got run times that aligned with your response, but couldn’t be sure that they were sufficiently general to draw conclusions.

David W