The Elvis problem revisited – Probably Overthinking It

[ad_1]

Right here’s an issue from Bayesian Data Analysis:

Elvis Presley had a twin brother (who died at beginning). What’s the chance that Elvis was an equivalent twin?

I’ll reply this query in three steps:

  • First, we’d like some background details about the relative frequencies of equivalent and fraternal twins.
  • Then we are going to use Bayes’s Theorem to bear in mind one piece of information, which is that Elvis’s twin was male.
  • Lastly, dwelling as much as the identify of this weblog, I’ll overthink the issue by making an allowance for a second piece of information, which is that Elvis’s twin died at beginning.

For background data, I’ll use knowledge from 1935, the 12 months Elvis was born, from the U.S. Census Bureau, Birth, Stillbirth, and Infant Mortality Statistics for the Continental United States, the Territory of Hawaii, the Virgin Islands 1935.

It contains this desk:

With a couple of affordable assumptions, we are able to use this knowledge to compute the chance that Elvis was an equivalent twin, provided that his twin brother died at beginning.

You can see my solution in this Jupyter notebook.

[ad_2]

Source link

Write a comment