Simple Random Sample -- How to Solve on TI-Nspire without Table?

[Luke Setzer]

A simple random sample will be obtained from a normally distributed
population. Find the minimum sample size needed to be 99% confident
that the sample standard deviation s is within 30% of sigma. Is such a
sample size practical in most cases?

Use a sample size determination table to determine the minimum sample
size needed to be 99% confident that s is within 30% of the true value
of sigma.

The values in the bottom right quadrant of the table are for 99%
confidence that s is within 30% of the true value of sigma.

The minimum sample size needed is 37.

Typical surveys involve sample sizes around 1000 to 2000.

[Luke Setzer]

I uploaded SampleSize.pdf to the Files folder of this group to convey
what I seek.

I want a closed form solution via the TI-Nspire CAS to supply these
numbers.

Can anyone help?

[Nelson Sousa]

for (almost) all due purposes, statistics on Nspire are the same as statistics on 83+ and 84+.

They do not compute confidence intervals for the standard deviation or variance (the PDF file you uploaded says that also), they only compute confidence intervals for the average.

The PDF mentions a program for the 83+/84+ that computes confidence intervals for sigma and sigma^2. You can translate it for the Nspire, as (in principle) all that runs on a 83+/84+ runs on Nspire (the same way as all that runs on 89/V200 runs on Nspire CAS), as long as the programs don't use graphics or any sort of fancy output.

If you want to know the capabilities of Nspire in terms of statistics I reccommend, besides Nspire's user guides, the 83+/84+ user guide that you can download from TI's website. Nspire doesn't do anything else (ok, maybe 1 little thing or two...).

For everything that Nspire can't do, if it's relevant there's probably some program available either for 89, V200 or 83+/84+ that does it, it's just a matter of looking into the right places (ticalc.org is a good place to start).

Hope this helps.

Nelson