Data Reduction and Error Analysis for the Physical Sciences, 2nd ed.

By P. R. Bevington, D. K. Robinson, G. Bunce. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Trend analysis of climate time series: A review of methods (2019, methodology): “See Bevington and Robinson (1992) for a practical view of the reduced sum of squares and the degrees of freedom.”
Vision-based global localization and mapping for mobile robots (2005, methodology): “For our global minimization, we can compute the covariance of the three scalar constraints from the uncertainty of each pairwise alignment based on first order error propagation [3]”
The Ionized Gas and Nuclear Environment in NGC 3783. I. Time-averaged 900 Kilosecond Chandra Grating Spectroscopy (2002, unknown): “The probability of observing Fχ ≥ 9.0 (3.6) for ν1 = 1 and ν2 = 48 dof is ∼ 0.5% (5%) (Table C5 of Bevington & Robinson 1992 ).”
Five New Extrasolar Planets (2005, methodology): “We thenfitaKeplerianmodel to the residual velocities, yielding its 2. The F-test determines the probability that one model constructed with additional free parameters is to be preferred over another model, as a function of the decrease in their respective 2 ( Bevington & Robinson 1992; A. Cumming 2”
FIT2D: a multi-purpose data reduction, analysis and visualization program (2016, background): “Arrays to store estimated variance values may be created, and error propagation performed assuming Gaussian error distributions (Bevington & Robinson, 1992).”