Metadata
- Author: Vicki Boykis
- Full Title:: The ghosts in the data
- Category:: 🗞️Articles, Data Analysis
- URL:: https://vickiboykis.com/2021/03/26/the-ghosts-in-the-data/
- Finished date:: 2021-05-03
Highlights
Real life phenomena, as I’ve seen them in industry, mostly do not follow a bell curve. Most of the data that people deal with across industries involves activities that people do: people on your platform, people clicking things, people receiving vaccines or medical treatments, restaurant visits, etc. And all of that data looks like this:
This essentially means that a few people do some activity a lot, and the rest not as much at all. Here are some real world examples: And much, much more. This means that we can’t deal with the data in the same way, particularly for smaller samples of data, because
Gaussian and Paretian distributions differ radically. The main feature of the Gaussian distribution … can be entirely characterized by its mean and variance … A Paretian distribution does not show a well-behaved mean or variance. A power law, therefore, has no average that can be assumed to represent the typical features of the distribution and no finite standard deviations upon which to base confidence intervals . . (View Highlight)
New highlights added 2024-10-15
The one thing that is never covered in any class is that you will have to deal with workplace politics, and how successful you are at navigating this hierarchy will determine how you’re perceived at work just as much as the actual quality of your work. (View Highlight)