モンティ・ホール問題好きのホームページ    privacy policy

Return to the list of my pages written in English about the two envelopes problem
2015/03/22 12:58:23
First edition 2015/01/01

Is the principle of insufficient reason the cause of the paradox of the two envelopes problem?

Caution
I who am Japanese wrote this page in English, but I am not so good at English.

The process through which the Two Envelope Paradox arise.

The process through which the Two Envelope Paradox (Exchange Paradox) arise is as follows.

The principle of insufficient reasen

The mechanism of this paradox is disappointingly simple.

We forget the odds of the pair (x/2 , x) , and the odds of the pair (x , 2x).
That why we thoughtlessly assign probability 1/2 to the terms in equation (1).
This is a kind of probability illusion called 'Base Rate Fallacy', and it is the cause of the two envelope paradox.

But some people don't think so.

They think that an improper application of  "the principle of insufficient reason"  is the cause why we assign a probability 1/2 to the event that on the condition the chosen envelope contains a specific amount of money, the chosen envelope contains lesser amount of money and why we assign a probability 1/2 to the event that on the same condition, it contains greater amount of money.

Comparison between 'base rate fallacy' and 'insufficient reason'

Terminology

I will use following terminology, in the following paragraphs and sections.

Majority

I think that the base rate fallacy is the cause of the paradox, but I have found no articles that say base rate fallacy is the cause of the paradox of the two envelopes problem. So my opinion is minor.

complexity

By a base rate fallacy, we simply confuse the likelihood and the conditional probability.
But to apply the principle of insufficient reason, we must think there is no information about the conditional probability except the likelihood.

consistency

I think that we can not unconsciously apply the principle of insufficient reason.
When we find a contradiction after we apply such a principle,  can we continue believing the probability?

If the base rate is described in the problem statement

I imagine a psychological experiment which use problem statement which describe following probability distribution.

pairs of amount of money probability
¥1,000, ¥2,000 1%
¥2,000, ¥4,000 49%
¥4,000, ¥8,000 49%
¥8,000, ¥16,000 1%

I expect that such a experiment will end with following result.

Return to the list of my pages written in English about the two envelopes problem