ANIMATING SAMPLES FROM GAUSSIAN DISTRIBUTIONS

You are here: Home / Submitted Papers / 2015 / ANIMATING SAMPLES FROM GAUSSIAN DISTRIBUTIONS

Abstract

The animations displayed below are animated samples from correlated Gaussian beliefs, following closed trajectories along equipotential lines of the probability distribution. They offer a more expressive view of the structure of samples from Gaussian processes than static samples. This document explains how to generate them, using Matlab, tikz, and LATEX, in that order. If you do not see two animations with wobbly lines on the first proper page of this document, try opening it with Adobe Reader.

Author

Philipp Hennig

Journal

Technical Note

Paper Publication Date

2013

Paper Type

Astrostatistics