Quantile-quantile plot for von Mises distribution fitted to circular data
circqvm varname [if exp] [in range] [, show(numlist) plot(plot) scatter_options]
circqvm gives a quantile-quantile plot for the fit of a von Mises (a.k.a. circular normal) distribution to a circular variable on a scale between 0 and 360 degrees. circvm is used to fit the distribution, estimating the two parameters vector mean mu and concentration parameter kappa. Both observed and expected quantiles are rotated so that each set is centred on the observed vector mean. Each is nevertheless labelled in terms of varname.
This plot was suggested by Fisher (1993). However, he uses a sine((angle - centre) / 2) scale, which has the merit of pulling in the tails and stretching the region near the centre of the distribution, but the disadvantage, as with any transformation, of being one step further away from the data.
show() specifies a numlist of axis labels to be shown, overriding the default.
plot(plot) provides a way to add other plots to the generated graph; see help plot_option.
scatter_options are any options allowed with scatter.
. circqvm wallasp . circqvm wallasp, show(0(45)315)
Fisher, N.I. 1993. Statistical analysis of circular data. Cambridge: Cambridge University Press.
Nicholas J. Cox, University of Durham, U.K. firstname.lastname@example.org
On-line: help for circvm, circpvm