Probability of flipping a biased coin k times

  • Thread starter Thread starter mcnkevin
  • Start date Start date
  • Tags Tags
    Probability
AI Thread Summary
The discussion revolves around calculating the probability of flipping a biased coin (with heads showing 2/3 of the time and tails 1/3) k times, specifically when k is odd, and determining the likelihood that tails appear more than heads. Participants initially misinterpret the question, leading to confusion about whether to compare tails to k or k/2. Clarifications reveal that the focus should be on the probability of tails exceeding heads, prompting further exploration of probability calculations for specific values of k. The conversation highlights the complexity of generalizing the solution and the need to consider independent tosses and various outcomes. Ultimately, the participants are working towards a clearer understanding of how to formulate the probability for k flips.
mcnkevin
Messages
6
Reaction score
0

Homework Statement


I have an unfair coin that comes ups heads 2/3 of the time and tails 1/3 of the time. If i flip this coin k times, what is probability that tails came up more then heads. Assume k is odd. So basically, what is the probability that the number of tails > k/2?

Homework Equations


The Attempt at a Solution



I got a mental block right now. Obviously the probability will depend on k, and the larger the the lower the probability. Is it correct if I say the probability is (1/3)^(k/2+1)?

Thanks
 
Last edited:
Physics news on Phys.org
Sorry if this is a stupid question, but how can tails come up more than the times you flip the coin?

If you flip it k times, it can't come up more than k times.
 
it says k/2 times not k.
 
mcnkevin said:
If i flip this coin k times, what is probability that tails came up more then k.

0 :wink:

Mezzlegasm was right, something is wrong here. Should it be

If i flip this coin k times, what is probability that tails came up more then k/2?
 
Oh right, yes. I think i actually meant to say in that sentance, what is the probability that it comes up more then heads. Later i mention k/2. Sorry.
 
mcnkevin said:
I have an unfair coin that comes ups heads 2/3 of the time and tails 1/3 of the time. If i flip this coin k times, what is probability that tails came up more then k.
Editorial note: This should say "what is probability that tails came up more then heads", not k. This is what led to Mezzlegasm's post.

Is it correct if I say the probability is (1/3)^(k/2+1)?
No. This isn't even true for the trivial case, k=1, in which the probability is obviously 1/3. What is the probability for k=3? What kind of distribution is this?
 
mcnkevin said:
Oh right, yes. I think i actually meant to say in that sentance, what is the probability that it comes up more then heads. Later i mention k/2. Sorry.

Yeah, I missed the latter part because I was confused with the former.

Anyhow, I'm pretty sure that it is (1/3)^(k/2). I always had a hard time with counting principles and probability, so I need someone to confirm that for me as well.

EDIT: alright, I've just seen the above post. My answer didn't seem to make sense because it gave a different probability with more coin tosses, which shouldn't happen (I don't think). I couldn't find a single way to make it work, and the probability that it comes up more than heads seems to be a more appropriate question. Regardless, my answer is flawed.
 
Not sure what you mean by what kind of distribution? Basically, each toss is independant and each toss you have 1/3 of a chance of flipping tails and 2/3 a chance flipping heads. If you make k flips, what's the probability tails was flipped more times then heads.

For k = 3 , well you have to flip 2 tails in 3 tries.

So the possibilities are
All three tails = (1/3)(1/3)(1/3)
First two tails = (1/3)(1/3)(2/3)
First and last tails (1/3)(2/3)(1/3)
Second and last tails = (2/3)(1/3)(1/3)

So basically, (1/3)^3 + (3*((1/3)^2)*(2/3))

Correct? It certainly may be very wrong.. but if it is correct, how could i generalize this to k?
 
mcnkevin said:
Not sure what you mean by what kind of distribution? Basically, each toss is independant and each toss you have 1/3 of a chance of flipping tails and 2/3 a chance flipping heads. If you make k flips, what's the probability tails was flipped more times then heads.

For k = 3 , well you have to flip 2 tails in 3 tries.

So the possibilities are
All three tails = (1/3)(1/3)(1/3)
First two tails = (1/3)(1/3)(2/3)
First and last tails (1/3)(2/3)(1/3)
Second and last tails = (2/3)(1/3)(1/3)

So basically, (1/3)^3 + (3(1/3)^2(2/3))

Correct? It certainly may be very wrong.. but if it is correct, how could i generalize this to k?

Ah alright, this is starting to feel familiar now.

Like you said there are different possibilities, so you have to add the different probabilities that have tails at the desired amount. I think this is in the right direction, someone correct me if I'm wrong. I'm not sure where to go now, but I believe it starts with

(1/3)^k +

then you would have to add the different possible probabilities when considering how many times you can flip heads and still have desirable results. I just don't know how to do this.
 
Back
Top