Math Is Fun Forum

  Discussion about math, puzzles, games and fun.   Useful symbols: ÷ × ½ √ ∞ ≠ ≤ ≥ ≈ ⇒ ± ∈ Δ θ ∴ ∑ ∫ • π ƒ -¹ ² ³ °

You are not logged in.

#1 2008-02-25 22:55:16

HP
Member
Registered: 2008-02-25
Posts: 2

Should I use Binomial or Poisson Distribution

I have a question with which i'm unusre on what method to answer it:

'You are a quality manager at a computer company that delivers 1500 computers each week. On average 2 computers fail to work on arrival

(i) Name a probability distribution that could be used to model this situation

(ii) This week 4 computers fail to work on arrival. What is the probability of 4 or more computers failing to work, calculated using the         distribution you have chosen'

I was leaning towards the decision of using poisson distribution because the question doesn't give an exact probability.

Am I correct in doing this?

Thanks

Offline

#2 2008-02-26 02:48:39

mathsyperson
Moderator
Registered: 2005-06-22
Posts: 4,900

Re: Should I use Binomial or Poisson Distribution

Yup, that's the model I'd choose.
Another way of thinking about it is that if you did go for a Binomial distribution, the calculations would be far too difficult, due to the large number of trials performed.

Even if a Binomial distribution was more suitable, it could be approximated accurately with Poisson, because of the high number of trials and low probability of success.


Why did the vector cross the road?
It wanted to be normal.

Offline

#3 2008-02-26 22:41:11

HP
Member
Registered: 2008-02-25
Posts: 2

Re: Should I use Binomial or Poisson Distribution

Hi again,

Thanks for the reply. I was just wondering, to work out the probability of 4 or more computers failing, would it be better to just work out the probability of 4 failing or should I calculate the probability of 0,1,2,3 computers failing?

Offline

#4 2008-02-27 03:13:21

mathsyperson
Moderator
Registered: 2005-06-22
Posts: 4,900

Re: Should I use Binomial or Poisson Distribution

Yes, you'd have to do it the way you described. You could work out the probability of exactly 4 computers failing directly, but that wouldn't answer the question.

You'd need to find the individual probabilities of 0, 1, 2 and 3 computers failing, and them all together and then take that away from 1.


Why did the vector cross the road?
It wanted to be normal.

Offline

Board footer

Powered by FluxBB