/biz/ - Business & Finance

Excel is literally used for everything, unless you need the matrix handling abilities of matlab/octave. The only real exception I can think of is using R for more intense experimental design work, but even then you can get away with Excel if you're wiling to load up like 10 different macros to do all the preliminary checks on a data set.

>>53124999
>>53125102
>>53125244
>>53125657
Powerfully good advice in this thread. Its actually kind of scary. What I can add to this, is that if you work in any sort of R&D capacity, or you even just deal with a problem where it could take a long time (or cost a lot) to try a bunch of different experiments/settings to see what works the best, you owe it to yourself to learn a little bit of basic experimental design, and the stats that go with it. Its a massive time save, and Fisher was a chad.

Last year, I had a coworker planning to try a bunch of different operational settings to determine which setup would lead to the maximum removal of metal pollutants in water. Each experiment would take about 3-4 days, and the total number of runs would be about 90 since he was looking at
>3 pH levels
>3 metal concentrations
>with additive / without additive
>5 flow rates

I stopped him and advised his team to first only try the extreme high and low of metal level and pH, the extremes of flow, and to try both additives. This would catch most of the variation in his experiments and only take 16 runs instead of 90, and then they could run finer experiments where they had the biggest responses on the response curve. I didn't even need real stats or design techniques, but it saved 82% of the time and effort it would have taken.

Who's your favorite billionaire?

To hijack a bit, what can go wrong If you just short Iron condor boomer ETFs?

>>52447060

>>52445543