Draws a normal quantile-quantile plot using symbols whose size is proportional to the probability that the associated observation was generated by the associated component of the model.
qqMix(object, xlim=NULL, ylim=NULL, shape="disc", ngon=20,
size=1, …)
A list with components containing the residuals, the relevant probabilities, the predictors, and the observations, as returned by residMix; probably the residuals should be standardized for use by qqMix, i.e. residMix should be called with std=TRUE.
The limits (endpoints) of the x axis of the plot; chosen in the "usual" way by the plotting software if xlim is NULL.
The limits (endpoints) of the y axis of the plot; chosen in the "usual" way by the plotting software if ylim is NULL.
The shape of the plotting symbol; possible values are "disc", "square", and "diamond". Partial matching is used so only the first few letters need be given.
The "disc" shape is actually a regular polygon; ngon specifies how many sides it should have; ignored if shape is not equal to "disc".
A scale factor to change the absolute sizes of the plotting symbols; values larger than 1 make the symbols larger than the default; values less than 1 make them smaller.
Additional arguments to be passed to the polygon function which actually draws the plotting symbols.
A normal quantile-quantile plot is drawn in whatever device is currently open.
The idea of creating quantile-quantile plots for regression mixtures by making the symbol size proportional to the associated probability is due to Adrian Baddeley of the University of Western Australia
The plot produced is visually assessed by ignoring or discounting small symbols.
Turner, T. R. (2000) Estimating the rate of spread of a viral infection of potato plants via mixtures of regressions. Appl. Statist. vol. 49, Part 3, pp. 371 -- 384.
bootcomp, cband, covmix, mixreg, plot.cband, plot.mresid, residMix
# NOT RUN {
#See \link{mixreg} for examples.
# }
Run the code above in your browser using DataLab