3 ms·
It's really not, I've used python in the past but prefer R for this. As it stands excel is a better presentation layer than almost all BI tools once you're pas
by nhinck3 2y ago
It's really not, I've used python in the past but prefer R for this.
As it stands excel is a better presentation layer than almost all BI tools once you're past the modelling and analysis stages.
- ttyprintk 2y agoWith this, you get: - Reproducibility (the official, bundled Excel numerical routines have/had errors greater than floating-point precision) which avoids the unprofessional look of, say, least squares numbers that differ from a check by hand. - Version and environment control. This is the fastest way I can answer the question, "what would these new routines produce if run against last October's pool of databases?" - A presentation format where client customizations for style, dimensional units and currencies, human language, etc. can all be owned outside of your project. I try to sell this approach when I can. Is there a particular BI that strikes a better balance?