For some reason that I don't fully understand, many developed nations like to monitor and publish their most popular baby names. The US, for example, keeps a complete list of the number of babies born with each name dating back to 1880. It's easy to produce cute visualisations with this stuff, but occasionally it can be useful too.
For example, using the data described above and a simple bit of statistical inference, it's possible to make a reasonable estimate the distribution of ages within a group, based only on their first names. I've stepped through the process in this jupyter notebook
- Reverse the process to guess the number of people with a certain name based on a distribution of ages.
- Weirdos also like aggregating popular dog names. Can I guess your dogs age based on its name?