The Student Room Group

Sampling distribution question

The random variable X has a B(40. 0.3) distribution. The mean of a random sample of n observations of X is denoted by x̄. Find(a) P(x̄≥13) when n is 49,(b) the smallest value of n for which P(x̄≥13)<0.001.
I've done part a and for b got upto (13)-(1 divided by 2n)-(2) all divided by (square root 8.4 divided by square root n)=3.090
How would i obtain a value for n?
Reply 1
for more clarity here's a picture
isnt xˉ(40,0.3n2) \displaystyle \bar{x} \sim (40, \sqrt{\frac{0.3}{n}}^2) and P(xˉ>X)=P(Z>X400.3n) \displaystyle P(\bar{x} > X) = P(Z > \frac{X - 40}{\sqrt\frac{{0.3}}{n}})

(just checking I have this right before I go any further)
Reply 3
Wouldn't the variance be 8.4 over n, since mean for binomial approximated to normal is npq, so 40*0.3*0.7=8.4
(edited 7 years ago)
Original post by gooner1010
Wouldn't the mean be 8.4 over n, since mean for binomial approximated to normal is npq, so 40*0.3*0.7=8.4


hmm as I feared, I cant remember how to do this :hide: sorry

Edit: Is this S2? a binomial to normal approximation question? ill revise it and get back to you if you want?
(edited 7 years ago)
Reply 5
Original post by DylanJ42
hmm as I feared, I cant remember how to do this :hide: sorry

Edit: Is this S2? a binomial to normal approximation question? ill revise it and get back to you if you want?


Hi, yep it's a binomial approximation to normal, and help would be appreciated.
Thanks
Original post by gooner1010
Hi, yep it's a binomial approximation to normal, and help would be appreciated.
Thanks


sorry I honestly dont have a clue how to do this. Hopefully someone else can help you
Original post by gooner1010
Hi, yep it's a binomial approximation to normal, and help would be appreciated.
Thanks


last try, is your answer to the first part 0.8869?

Spoiler

Reply 8
Original post by DylanJ42
last try, is your answer to the first part 0.8869?

Spoiler




Sorry. it's 0.0084
Original post by gooner1010
Sorry. it's 0.0084


ive got no idea then, sorry :laugh:

iirc @SeanFM knows his stats, maybe he can salvage this
Original post by gooner1010
The random variable X has a B(40. 0.3) distribution. The mean of a random sample of n observations of X is denoted by x̄. Find(a) P(x̄≥13) when n is 49,(b) the smallest value of n for which P(x̄≥13)<0.001.
I've done part a and for b got upto (13)-(1 divided by 2n)-(2) all divided by (square root 8.4 divided by square root n)=3.090
How would i obtain a value for n?


So this is a normal approximation to the binomial type question. If you have a random variable X distributed as B(n,p) then it is approximated by a variable distributed as N(np, np(1-p)) (where the second parameter for the normal is given as the variance rather that standard deviation).

So here, X is approximated by N(40 * 0.3, 40 * 0.3 * 0.7) = N(12, 8.4)

If X is distributed as N(μ,σ2)N(\mu, \sigma^2), then X-bar is distributed as N(μ,σ2/n)N(\mu, \sigma^2/n) where n is the number of observations. So here we have the mean distributed as N(12, 0.1714286). So part (a) follows by looking at tables or using software.

If n is unknown, then you need to work in reverse, try a few values of n and see what values you get for the probability - after a while you should be able to pinpoint the n you want.
Reply 11
Original post by Gregorius
So this is a normal approximation to the binomial type question. If you have a random variable X distributed as B(n,p) then it is approximated by a variable distributed as N(np, np(1-p)) (where the second parameter for the normal is given as the variance rather that standard deviation).

So here, X is approximated by N(40 * 0.3, 40 * 0.3 * 0.7) = N(12, 8.4)

If X is distributed as N(μ,σ2)N(\mu, \sigma^2), then X-bar is distributed as N(μ,σ2/n)N(\mu, \sigma^2/n) where n is the number of observations. So here we have the mean distributed as N(12, 0.1714286). So part (a) follows by looking at tables or using software.

If n is unknown, then you need to work in reverse, try a few values of n and see what values you get for the probability - after a while you should be able to pinpoint the n you want.


The book has the answer 82, however I seem to be getting 81
Original post by gooner1010
The book has the answer 82, however I seem to be getting 81


I agree with your value of 81. I'm using the following R code

N <- 81
mm <- 12
ss <- (40 * 0.3 * 0.7)/N
1 - pnorm(13, mm, sqrt(ss))

Quick Reply

Latest