pyerrors: A python framework for error analysis of Monte Carlo data

Published: 8 May 2023| Version 1 | DOI: 10.17632/7ncw242ymh.1


We present the pyerrors python package for statistical error analysis of Monte Carlo data. Linear error propagation using automatic differentiation in an object oriented framework is combined with the Γ-method for a reliable estimation of autocorrelation times. Data from different sources can easily be combined, keeping the information on the origin of error components intact throughout the analysis. pyerrors can be smoothly integrated into the existing scientific python ecosystem which allows for efficient and compact analyses.



Computational Physics, Error Analysis, Markov Chain, Application of Monte Carlo Method