Asymmetric Statistical Errors

Abstract

Asymmetric statistical errors arise for experimental results obtained by Maximum Likelihood estimation, in cases where the number of results is finite and the log likelihood function is not a symmetric parabola. This note discusses how separate asymmetric errors on a single result should be combined, and how several results with asymmetric errors should be combined to give an overall measurement. In the process it considers several methods for parametrising curves that are approximately parabolic.

14 pages, 5 figures

References in corpus (2)

• Likelihood inference in the presence of nuisance parameters
• Averaging Measurements with Hidden Correlations and Asymmetric Errors

Cited by in corpus (14)

• Measurement of b hadron production fractions in 7 TeV pp collisions
• Excess AGN Activity in the z=2.30 Protocluster in HS 1700+64
• Coherent $\psi$(2S) photo-production in ultra-peripheral Pb-Pb collisions at ${\sqrt{s}}_{\mathrm{NN}}$ = 2.76 TeV
• Publishing statistical models: Getting the most out of particle physics experiments
• The Eddington ratio-dependent changing look events in NGC 2992
• Reproducibility and Replication of Experimental Particle Physics Results
• Flavour specific neutrino self-interaction: $H_0$ tension and IceCube
• A spectroscopic test of the rotational modulation origin of periodic \emph{Kepler} photometric variability of A-type stars
• The shapes and alignments of the satellites of the Milky Way and Andromeda
• Microwave spectroscopic study of the hyperfine structure of antiprotonic helium-3
• Masgomas-4: Physical characterization of a double-core obscured cluster with a massive and very young stellar population
• Combining parameter values or $p$-values
• Towards a precise measurement of the cosmic-ray positron fraction
• Probing non-thermal light DM with structure formation and $N_{\mathrm{eff}}$