The Optimal Stopping Theorem
The optimal stopping theorem is a problem that is very famous in the mathematical field. In
technical terms it is known as “choosing a time to take a given action based on
sequentially observed random variables in order to maximize an expected payoff or to
minimize an expected cost”. In simpler words it is the simple problem we face a lot in our
life – When should we stop.
There are many variations to problems that apply the optimal stopping theorem, and we
can use them in real life. One such variation that can be used is called the secretary
problem. In this problem you are the employer and it is your job to employ the best
candidate. You get to interview the candidate in a random order, and if you reject them
they will instantly go away, and you can not employ them anymore, but if you accept them
they will instantly accept the job. Now the question is what will be your strategy to
maximize your chances to employ the best candidate. The more you think about this
question the more harder it seems to get, will you employ a good candidate as soon as you
see one, or will you wait in hopes of a better candidate when there might not be one.
The math behind such a problem is very complex, if you would like to understand it better I
do suggest you look into it. Now I will explain the best strategy for this problem – The look
then leap strategy. In this strategy the first few applicants will be the ones you just interview
and not accept any one no matter how good they may be. Instead we will use this time to
set up a ranking of the applicants we have seen so far. Once the set number of applicants
have passed we will now begin the “leap” phase of the plan, in which we will immediately
hire the next best applicant in relation to our ranking.
Now that we have determined
a strategy to tackle the
problem, let us now investigate
the next natural question that
arises: How many applicants
should be in our looking
period? The math again is too
complex for me to explain here,
but there is a definitive answer.
The answer is that you should look through (100/e)%, where e is the mathematical constant.
This approximates to 37%, which is the generally agreed percentage. This means that you
should look through approximately 1/3rd of the applicants before you immediately go for the
next best one.
Great, now we have the best possible strategy for this situation, so now we are prepared for
this if such a scenario ever comes up in real life, however another important fact that I’m
sure you are all thinking about is what are the chances of success with such a strategy.
Seeing as it is the best possible strategy out there, maybe you would be expecting
something high like 70%, but you would actually be way off. The chances of success are yet
again 37% !. This is another problem in math that seems to have this very nice symmetry. I
bet you are all very disappointed that this is the best possible outcome that we could
produce, and you wouldn’t be wrong to feel this way, however such is expected with the
constraints of the question presented.
You may now be thinking, great now I know the solution to this random niche problem that I
will probably never occur in real life. Well I would like to say that you are completely wrong.
This strategy can be applied to many problems in the real world, two of the main ones
being: buying a house, where you spend 37% of the time allotted to buy the house just
looking around, and maybe one that is more relatable, Dating.
Such a situation is directly related to dating and thus mathematics advises that you spend
your youth rejecting everyone to get an idea of the field and making a ranking, so that when
you use up 37% of your allotted time, you have a better chance to find your “Fated partner”.
I am however unfortunately not an expert on this topic, but for a more in depth solution I
suggest reading the article “When should you settle down ?”, by Hannah Fry. An interesting
read where love and math collide, to create unique equations such as:
Where P is the probability that you settle down with the best person you will ever meet, n is
the number of potential lovers you have, and r is the number of them you reject.
I hope that this article gave you some good insight as to how you could use math in real life
to have the best chances to make the best choices in life, and also as to how powerful a tool
maths really is. The implications of the subject extend far beyond the classroom problems or
finding the area of some cuboid or finding x in a simultaneous equation. These are but tools
that you require to define the world in terms of math, and once you do, your capabilities
expand. Thank you for reading, I do hope it was enjoyable.
- By Jidneya Desale
