Hi all,
I need to find methods for extrapolating model accuracy results – I hope you can help me. To put this into context, say, for example, we have a sample size of 100,000 people, and we scan a piece of information they enter to verify & identify their identities drawing on information from a database. What we then want to do is draw on the accuracy results (e.g. 98% True Match Rate / 2% False Match Rate) from this sample and extrapolate this to a much larger sample size (e.g. 100,000,000 people) so that we can know how accurate the model will be for such a sample size (by considering Type I & II errors, etc). Does anyone know which methods I can use to do this?
Thanks.