4 ms·
1. I only ever heard of this as a dilemma and not a paradox 2. There is this recruiting problem where you have to decide which secretary to hire before having
by dmichulke 9y ago
1. I only ever heard of this as a dilemma and not a paradox
2. There is this recruiting problem where you have to decide which secretary to hire before having seen all of them. The optimal solution is to see the first 37% (1/e) and then hire the first that is better than the previously seen best.
Putting that into lifespan of ~75 ys and assuming we don't really start sampling until we are 15, we end up with an optimal stopping (= exploitation) age of 32.07.
Surprisingly close to the 30 mentioned.
https://en.wikipedia.org/wiki/Secretary_problem https://en.wikipedia.org/wiki/Secretary_problem
- thecrazyone 9y agoOh! I used a similar mathematical model while finding a spouse. I figured, that for me, 37% (1/e) worked out to 13 people. This (finding people) is a space I've spent a lot of time doing, planning, understanding, and thinking about :) Also, there are a lot of scientific papers on how to model these in scholar.google.com (sorry that I don't have links that I can give right now)
- TorKlingberg 9y agoIt's the optimal solution if you wan to optimize the probability of hiring the one best secretary out of all the candidates. When hiring you usually care more about maximizing the expected "goodness", hiring someone in the top X%, or anyone who meets a certain bar.