Astrostatistics Tools (astropy.stats)

Introduction

The astropy.stats package holds statistical functions or algorithms used in astronomy and astropy.

Getting Started

The current tools are fairly self-contained, and include relevant examples in their docstrings.

See Also

  • scipy.stats

    This scipy package contains a variety of useful statistical functions and classes. The functionality in astropy.stats is intended to supplement this, not replace it.

Reference/API

astropy.stats Package

This subpackage contains statistical tools provided for or used by Astropy.

While the scipy.stats package contains a wide range of statistical tools, it is a general-purpose package, and is missing some that are particularly useful to astronomy or are used in an atypical way in astronomy. This package is intended to provide such functionality, but not to replace scipy.stats if its implementation satisfies astronomers’ needs.

Functions

binned_binom_proportion(x, success[, bins, ...]) Binomial proportion and confidence interval in bins of a continuous variable x.
binom_conf_interval(k, n[, conf, interval]) Binomial proportion confidence interval given k successes, n trials.
biweight_location(a[, c, M]) Compute the biweight location for an array.
biweight_midvariance(a[, c, M]) Compute the biweight midvariance for an array.
bootstrap(data[, bootnum, samples, bootfunc]) Performs bootstrap resampling on numpy arrays.
mad_std(data) Calculate a robust standard deviation using the median absolute deviation (MAD).
median_absolute_deviation(a[, axis]) Compute the median absolute deviation.
sigma_clip(data[, sig, iters, cenfunc, ...]) Perform sigma-clipping on the provided data.
sigma_clipped_stats(data[, mask, mask_val, ...]) Calculate sigma-clipped statistics from data.
signal_to_noise_oir_ccd(t, source_eps, ...) Computes the signal to noise ratio for source being observed in the optical/IR using a CCD.