Can you elaborate a bit more on “monthly challenges” and how this relates? Are you trying to change specific behaviours and using correlations to understand whether your changes have an effect?

For the monthly challenge, it could be things like changing diet, sleep habits, exercise, etc.

I track food only a few months of the year to get an idea of what’s in the food I eat, and sporadically the rest of the time. I don’t want incomplete data to affect correlations when I’ve found motivation to start tracking everything thoroughly. I don’t want it telling me “you are more active when you consume less sugar,” only because I wasn’t tracking sugar before.

The same goes for exercise. If I find motivation to start a challenge of exercising every day in December, I don’t like having stale data from 11 months ago affecting my correlations. If I started exercising more and my productivity increased immediately, I would want to know about that strong correlation instead of a very slight increase when averaged over the year.

Part of the personal challenge idea is to improve yourself, but I don’t like missing data or past failures skewing correlations during a new challenge.

Also, similar to the Eras suggestion, major life changes can make old data not relevant now.

“you are more active when you consume less sugar” because I wasn’t tracking sugar before

We ignore days with N/A data for correlations, so something like this isn’t possible. But I can still see why you might want to compare only a specific period to see if the correlation is stronger or weaker, thanks for the clarification 🙂