[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Sample size for equivalence trials in Stata
Joseph Coveney wrote:
> It seems as if Philip Ryan's program would work with Stata 8.2. Did you
> take note of the erratum
> ( www.stata.com/statalist/archive/2003-02/msg00212.html ) that he posted
> shortly afterward?
I typed an error when I corrected the ado-file... sorry. Now equivsize works,
but it allow for sample size calculation with binary outcome and I have a
continuous outcome (mean number of events).
> Regardless of whether you're working with proportions or another data type,
> simulation would be an option. See the references Philip give in this
> regard in www.stata.com/statalist/archive/2003-02/msg00211.html . The
> basic algorithm for proportions, for example, would be to generate a few
> thousand sets of data at the boundary of the alternative hypotheses--that
> is, at your chosen delta--using Joseph Hilbe's and Walter Linde-Zwirble's
> -rnd-, to calculate confidence intervals for the risk difference (risk
> ratio or odds ratio, whatever), and then to determine the proportion of the
> datasets in which the confidence intervals contained zero (or one).
This is very interesting, but I don't know how to it... could you help me with
* For searches and help try: