What is the expected number of flips to observe the first 5 flips pattern again?

  • Context: Graduate 
  • Thread starter Thread starter mmehdi
  • Start date Start date
Click For Summary

Discussion Overview

The discussion revolves around calculating the expected number of coin flips required to observe a specific pattern of five flips again, including the possibility of using some of the initial flips. The scope includes theoretical exploration and mathematical reasoning related to probability and expected values.

Discussion Character

  • Exploratory
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant suggests that the expected number of additional flips after the initial five flips is key to solving the problem, but finds it complex to derive a general solution for all combinations of five flips.
  • Another participant proposes using a linear difference equation for auxiliary probabilities, indicating a computational approach might be necessary to solve the problem.
  • A different viewpoint considers that each flip after the initial five creates a new sequence of five flips, questioning whether the situation can be modeled as a Bernoulli trial or through an indicator function variable.
  • One participant describes specific conditions under which the sequence would terminate, providing examples of sequences that would lead to termination at various flip counts.
  • Another participant references the problem's origin from an old exam and discusses a method involving probabilities of termination at different flip counts, expressing difficulty in extending this to an infinite series.

Areas of Agreement / Disagreement

Participants express various methods and approaches to tackle the problem, but there is no consensus on a definitive solution or methodology. Multiple competing views and uncertainties remain regarding the best way to calculate the expected number of flips.

Contextual Notes

Participants highlight challenges in deriving a general solution, including the complexity of the problem and the need for potentially infinite series calculations. There are also assumptions about the independence of flips and the nature of the sequences involved.

mmehdi
Messages
6
Reaction score
0
A coin having probability p of landing heads is flipped 5 times and then flipped utill the pattern of the first 5 flips is observed again for the first time, including the possiblity of using some of the first flips. If we want to find the expected number of flips of coins.(The lead is that: The expected number of additional flips after the five flips shall give us the answer)

...I have kind of tried to do it a couple of ways, but a solution which holds for all combination of five flips seems is proving to be too confounding.
 
Physics news on Phys.org
Eek, where did this problem come from? The only method I see for solving it is setting up a giant linear difference equation for some auxiliary probabilities, thus allowing me to use a computer algebra package to solve the system and manipulate it to get the answer you want.
 
After the first five flips, every single flip shall lead to a new sequence of five flips. So the probability that the sequence shall terminate is whether every flip is such that it regenerates the earlier sequence. So I wonder if it could be turned into a burnoulli trial, or the expectation of an indicator function variable.

I have an intuitive feeling that this should have a rather straight forward close form solution.
 
Expected Number of flips

In order for the sequence to terminate on the sixth flip. The first five will have to be all heads or all tails. Meaning all the flips should be equal to the first flip. For the sequence to terminate in 7 flips the sequence should be (12121). Third flip equal to first flip, fourth equal to 2nd, fifth equal to 3rd, sixth equal to 4th and seventh equal to 5th.

For the sequence to terminate on the 8th flip the sequence will have to be (12312), 9 th flip (12341), and tenth flip (12345).
 
The problem came in an old exam, three years ago. One of the way to do it, is E[X]=P[X>1]+P[X>2]+P[X>3]...

Where P[X>1]= P[1-P[X=1]. Where P[X=1] which means termination at one additional flip is p^6(1-p)^6. I am just having trouble in extending it to infinite series
 

Similar threads

Replies
4
Views
3K
  • · Replies 2 ·
Replies
2
Views
5K
  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 45 ·
2
Replies
45
Views
6K
  • · Replies 2 ·
Replies
2
Views
5K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 18 ·
Replies
18
Views
13K
  • · Replies 6 ·
Replies
6
Views
10K
  • · Replies 15 ·
Replies
15
Views
8K