Premium
Taylor's Power Law for Dependence of Variance on Mean in Animal Populations
Author(s) -
Perry J. N.
Publication year - 1981
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2346349
Subject(s) - variance (accounting) , mathematics , statistics , law , econometrics , political science , economics , accounting
S ummary Taylor (1961) suggested that population variance is proportional to a power of population mean for counts of animals sampled simultaneously at several sites. Three models which enable estimation of the exponent in this relationship are examined. Each is an empirical version of Taylor's law with population moments replaced by sample statistics. Some conditions are derived for satisfactory estimation when these models are used. Certain problems in estimation are examined; the practical severity of these vary between models. Methods of assessing these problems are developed for use with any data set, and the models are examined using these methods for a large set of moth data. The model of Taylor (1961) performed fairly well, and should prove satisfactory for use with similar sets of animal data.