MACSS

Part 6: Going from per-object estimates to ensemble estimates

Goals: go from per-object estimates to estimates of the distribution of p(z) for an ensemble of objects.

Specifics: Up to this point we have been worrying about the redshifts of individual objects. There are a lot of times what we care about is actually the distribution of the redshifts of a great many objects. We typically call these n(z) distributions, as opposed to per-object p(z).

You will want to:

1. come up with a naive distribution of the n(z) for object in the test sample
2. apply the same methodology to a sample of object that we do not have reference redshifts for.
3. consider the systematic different between the test sample and the broader sample and how we might account for those.

This site is open source. Improve this page.