A Data Science Central Community
Well, here are some stupid math tricks to keep you occupied for a while. They can be elegant tricks from random samples or brute force, either way, they avoid all the acrobatics in determining the shape of a distribution and going to an obscure textbook to find the best measure of central tendency. Here at the dashboard factory, we hear certain assertions as to the claim that the median is always the best or that the mean is always the best.
In spreadsheet models that get passed from one well meaning analyst to another, as we all know, the error of these spreadsheet outcomes becomes more and more exponential as people refuse to acknowledge the looming catastrophic outcome that can bring a really big organization to its knees.
Just ask a certain London based financial institution for more info on their denial of this fact in regards to the copy and paste mentality.
Unfortunately, BI packages do not really address this solution, other than the copy and paste and audit issues, because BI packages just leave audit trails and ensure that everyone is working from the worst information possible.
What is the solution to our rules of thumb? If only there were some buzzword laden stupid math tricks to the rescue.