Glad to receive a reply to the problem Glen. Well the observations from
two samples are independent.

So far I have progressed to :

I want to test the difference between the means of the two samples which
come from bernoulli populations. (the random variable used to generate the
population is a coin toss with 1 or 0) . The mean of a bernoulli sample is
essentially an estimate for the probability of 1's .

I approximated that the two populations are normal. Then, I ran the t-test
for difference in means of independent samples and get decent but not
highly satisfactory results. My sample sizes are decently large about >
50, so running fisher's exact test is not required I believe. Since I
still havent studied about chi-square, I do not know abt it.

I am still searching if there exists a good test which can test the
difference between means of two bernoulli populations.

The question is -- Is the difference between means of two samples which
come from bernoulli populations going to be normally distributed ?

If yes, then it may be right to use t-test, if not t-test is an
approximation  ?? .

regards,
vijay

-------------------------------------------

On 28 Apr 2004, Glen wrote:

> Vijay Arya <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>...
> > Hello,
> >
> > I have the following problem.
> >
> > -----------------------------
> >
> > A user T has a coin with head or tails and we want to estimate the
> > probability of head.
> >
> > when user A goes to T, he tosses the coin Na times giving a binary 1010...
> > trace (1=heads, 0=tails).
> >
> > when user B goes to T, he tosses the coin Nb times giving a binary
> > 1010...trace (1=heads, 0=tails).
> >
> > Now,
> >
> > A takes the mean Mua = (1+0+1+...)/Na  (essentially the probability of
> > heads)
> >
> > B does the same, Mub = (1+0+1....)/Nb
> >
> > I want to test the hypothesis
> >
> > H0: Mua not equals Mub
> > H1: Mua equals Mub
>
> Your hypotheses are the wrong way around.
>
> > I wanted to know if this falls under "inference from two
> > dependent or independent samples"
>
> Can you describe a way in which you think that the observations
> from the two samples might be dependent?
>
> Glen
>

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to