Optimal company selection (stopping) time

searcher profile

December 13, 2020

by a searcher from Princeton University in San Francisco, CA, USA


The problem of choosing deals when you come across them sequentially is well studied and has applications in many fields - I am sharing links here because someone starting out may find it useful to think about their process from this lens.

https://en.wikipedia.org/wiki/Odds_algorithm

https://en.wikipedia.org/wiki/Secretary_problem

3
4
183
Replies
4
commentor profile
Reply by a searcher
from Massachusetts Institute of Technology in New York, NY, USA
My understanding is that the "37% algorithm" applies when you can't go back to a deal you've passed. I don't think that's the case in a search - maybe the seller is just not ready right now (esp in non-brokerd deals). Plus, you realistically look at multiple deals concurrently over some time. I'm sure there is a mathematical formulation that reflects these changes in constraints but I bet the optimal answer would look different
commentor profile
Reply by a searcher
from École Centrale Paris in Paris, France
Really interesting conceptually, a lot of questions if applicable to a search....

But, if it were applied to the Search, maybe we could assume that deal flow is linear over time and we could assume that it takes some time to start seeing good deal flow, let's say 6 months, and after that you can apply the 1/e rule. So for a 24 month search, it would be the most interesting deal after 6+(24-6)/e = ~13 months of search.
commentor profile
+2 more replies.
Join the discussion