This documentation is for astroML version 0.2

This page


astroML Mailing List

GitHub Issue Tracker


Scipy 2012 (15 minute talk)

Scipy 2013 (20 minute talk)


If you use the software, please consider citing astroML.

11.6.7. astroML.stats.fit_bivariate_normal

astroML.stats.fit_bivariate_normal(x, y, robust=False)

Fit bivariate normal parameters to a 2D distribution of points

Parameters :

x, y : array_like

The x, y coordinates of the points

robust : boolean (optional, default=False)

If True, then use rank-based statistics which are robust to outliers Otherwise, use mean/std statistics which are not robust

Returns :

mu : tuple

(x, y) location of the best-fit bivariate normal

sigma_1, sigma_2 : float

The best-fit gaussian widths in the uncorrelated frame

alpha : float

The rotation angle in radians of the uncorrelated frame