logarithmic distribution


Also found in: Wikipedia.

logarithmic distribution

[′läg·ə‚rith·mik ‚dis·trə′byü·shən]
(statistics)
A frequency distribution whose value at any integer n = 1, 2,… is λ n /(-n) log (1-λ), where λ is fixed.
McGraw-Hill Dictionary of Scientific & Technical Terms, 6E, Copyright © 2003 by The McGraw-Hill Companies, Inc.
Mentioned in ?
References in periodicals archive ?
(2) Let [a.sub.n] = [p.sup.n]/n, where 0 < p < 1, n [member of] N = {1,2,3, ...}; then it reduces to Logarithmic Distribution [1].
Contaminant distributions follow a logarithmic distribution, and, therefore, analyses were done on geometric means.
By using the least square method and chi-square test to fit headway distribution, headway distribution conforms to logarithmic distribution no matter whether the average value or headway value at any quintile is used.
presented a quantitative method of investigating curves used in automobile design which is called Logarithmic Distribution Diagram of Curvature (LDDC).
Profiles of measurements far from the cylinder have the logarithmic distribution, whereas close to the cylinder, from P (15.0, 0[degrees], z) to P (12.0, 0[degrees], z) and from Q (15.0, 0[degrees], z) to Q (12.0, 0[degrees], z), the shape of [bar.u](z)/[U.sub.[infinity]] becomes more rectangular, and the maximum velocity is already reached far away under the water surface.
Even though the distribution of species abundance in ditches and paths did not fit any of the series models, it came close to being within the limits of the logarithmic distribution model (Table II).
Most seismic hazard software (Bender and Perkins, 1987; Ordaz et al., 2003) is designed to work with PGA or spectral acceleration such as residuals following a normal logarithmic distribution (Sabetta and Pugliese, 1987; Ambraseys et al., 1996; Boore et al., 1997).
The computer code adopted to calculate the seismic hazard was SeisRisk III (Bender and Perkins, 1987), which has been modified to be used with macroseismic intensity data (e.g., allowing a normal distribution instead of a normal logarithmic distribution of the residuals).
For the load-dependent model, Poisson, Binomial Negative, Positive Binomial, and Logarithmic distributions of [Lambda] are considered successively.
The Negative Binomial and Logarithmic distributions differ from the two other distributions in that they are themselves overdispersed.