Come analyze HEASARC, IRSA, and MAST data in the cloud! The Fornax Initiative is now welcoming all interested beta users.
A worked example: a simulated Chandra spectrum
To make the trade-offs concrete, we walk through the same fit with all four samplers on a single inexpensive problem. The data is a fakeit Chandra ACIS-S spectrum (RMF and ARF taken from cycle 16, 512 channels in the 0.5–8 keV band) with a 10 ks exposure, fit to TBabs*apec with three thawed parameters (nH, kT, norm; abundance held at solar). Numbers given below are from a single dual-cycle run of the XSPEC benchmark harness on a four-core Apple Silicon host; absolute elapsed times will differ on other hardware but the relative ordering is robust.
Subsections
- Setting up the data and finding the MLE
- Running each sampler
- Comparing the results
- When to pick what, for this kind of problem