Graph effects of individual circular values on vector mean and vector strength
circovmean varname [if exp] [in range] [, nolist circscatter_options ]
circovstr varname [if exp] [in range] [, nolist circscatter_options ]
Description
circovmean and circovstr list and graph the effects of omitting individual values on the vector mean and vector strength respectively of a circular variable measured on a scale between 0 and 360 degrees. For more information on these measures, see circsummarize.
Remarks
Showing the effects of omitting individual circular values on summary measures is a simple way of screening for outliers and/or assessing the importance of apparent or obvious outliers. Barnett and Lewis (1994, Ch.11) supply more discussion and references on working with possible circular outliers. One of their examples concerns homing directions, measured as degrees from North) of 14 northern cricket frogs Acris crepitans (Ferguson, Landreth and McKeown 1967; Collett 1980): 104, 110, 117, 121, 127, 130, 136, 145, 152, 178, 184, 192, 200, 316.
Options
nolist suppresses the listing of values.
circscatter_options are options of circscatter.
Examples
. circovmean axisasp
Author
Nicholas J. Cox, University of Durham, U.K. n.j.cox@durham.ac.uk
References
Barnett, V. and Lewis, T. 1994. Outliers in statistical data. Chichester: John Wiley. (Previous editions: 1978, 1984.)
Collett, D. 1980. Outliers in circular data. Applied Statistics 29: 50--57.
Ferguson, D.E., Landreth, H.F. and McKeown, J.P. 1967. Sun compass orientation of the northern cricket frog, Acris crepitans. Animal Behaviour 15: 45--53.
Also see
On-line: help for circsummarize, circmedian, circscatter