Bayesian Low-Counts X-ray Spectral Analysis

| Code | References | Usage | How to |

[under construction]
Last Updated: 2008apr03

BLoCXS is a Bayesian code that uses Markov-Chain Monte Carlo and Data Augmentation techniques to analyze CCD-resolution X-ray spectra in the Poisson regime. BLoCXS can be used to
  1. fit common emission models to low-counts spectra
  2. identify narrow emission lines in low-resolution spectra
  3. identify Gaussian emission lines (with fixed/unknown location and width)
  4. compute profile posterior distributions,
  5. simulate data from posterior predictive distributions,
  6. compute the maximum likelihood using posterior predictive draws (to calculate ppp-values)
  7. incorporate effective areas calibration uncertainty

The code

The code is in alpha. Contact the authors if you want to try it. The structure of the algorithm is represented by the data flow diagram shown below:


Protassov, R., van Dyk, D.A., Connors, A., Kashyap, V.L., & Siemiginowska, A., 2002, ApJ, 571, 545
van Dyk, D.A., Connors, A., Kashyap, V.L., & Siemiginowska, A., 2001, ApJ, 548, 224

Usage and Syntax

Under construction

How To

Under construction

| Code | References | Usage | How to |