China
Africa
Explore chapters and articles related to this topic
Drilling and Rock Mechanics
Published in C.P. Chugh, Ken Steele, V.M. Sharma, Design Criteria for Drill Rigs: Equipment and Drilling Techniques, 2020
C.P. Chugh, Ken Steele, V.M. Sharma
Sample statistics deal with relations existing between a population and samples drawn from the population. For conclusions of the sampling theory and statistical inference to be valid, samples must be so chosen as to be representative of a population. In statistics Chauvenet’s criterion, a special case of the t-distribution, is used in selecting the observations which belong to a population and rejecting outlier observations. The criterion states: ‘An observation in a sample of size N is rejected if it has a deviation from the mean greater than that corresponding to a 1/2 N probability. The criterion is discussed in some detail elsewhere; rules for use of the criterion are given below (after Goktan and Ayday, 1993). Compute the mean and standard deviation of all observations.Determine the ratio of the ‘suspiciously’ large deviation divided by the standard deviation. Determine the limiting value P of this ratio from Table 6.1, for the corresponding number of determinations N. If the observed ratio is greater than the value found in the table, the observation may be rejected.
Statistics
Published in Benjamin D. Shaw, Uncertainty Analysis of Experimental Data with R, 2017
Chauvenet’s criterion states that a data point can be considered to be an outlier if it is outside a certain probability band around the data mean. In using this criterion, we assume that the data are normally distributed and the criterion should only be used once; i.e., if outliers are rejected from a data set, we do not use Chauvenet’s criterion to identify and reject any more outliers from the remaining data.
Comparison of Measurement Reliability of Nanosecond Rectangular Voltage Pulses by Kerr Effect and by High-Speed Voltage Probe
Published in Fusion Science and Technology, 2022
Nemanja Aranđelović, Dušan Nikezić, Dragan Brajović, Uzahir Ramadani
Statistical samples of the 100 random variables “phase difference” and “pulse voltage” were processed statistically by the following steps: Applying Chauvenet’s criterion, suspicious measurement values are discarded.Statistical distributions to which random variables belong are determined.The first three central moments of the statistical distributions of random variables are determined.The measurement uncertainty type A of statistical samples of random variables was determined.
Evidence-based assessment of energy performance of two large centrifugal chillers over nine cooling seasons
Published in Science and Technology for the Built Environment, 2021
The outliers in these data sets are eliminated using Chauvenet’s criterion (Equation (7)) (ASHRAE 2010). Any measured value m of a variable x is an outlier, when its deviation (m-xmean) from the mean value (xmean) of dataset exceeds the maximum difference dmax: where n is the dataset size, and S is the standard deviation in the data set.