isoPortionLines {compositions} | R Documentation |
Add lines of equal portion and proportion to ternary diagrams.
isoPortionLines(...) isoPortionLines.acomp(by=0.2,at=seq(0,1,by=by),...,parts=1:3,total=1,labs=TRUE,lines=TRUE,unit="") isoPortionLines.rcomp(by=0.2,at=seq(0,1,by=by),...,parts=1:3,total=1,labs=TRUE,lines=TRUE,unit="") isoProportionLines(...) isoProportionLines.acomp(by=0.2,at=seq(0,1,by=by),...,parts=1:3,labs=TRUE,lines=TRUE) isoProportionLines.rcomp(by=0.2,at=seq(0,1,by=by),...,parts=1:3,labs=TRUE,lines=TRUE)
... |
graphical arguments |
at |
numeric in [0,1]: Which portions/proportions should be marked? |
by |
numeric in (0,1]:Steps between protions/proportions |
parts |
numeric vector subset of {1,2,3}: the variables to be marked |
total |
The total amount to be used in labeling |
labs |
logical: plot the labels? |
lines |
logical: plot the lines? |
unit |
marking the units e.g. "%" |
Isoportion lines give lines of the same portion of one of the parts, while isoproportion line gives lines of the same ratio between two parts. The isoproportion lines are Aitchison-lines in the simplex, while the isoportion-lines are, up to my knowledge, not.
Currently IsoLines only work individual plots. This is mainly due to the fact that I have no idea, what the user interface of this function should look like for multipanel plots. This includes philosophical problems with the meaning of isoportions in case of marginal plots.
K.Gerald v.d. Boogaart http://www.stat.boogaart.de
data(SimulatedAmounts) plot(acomp(sa.lognormals)) isoPortionLines() plot(acomp(sa.lognormals),center=TRUE) isoPortionLines() plot(rcomp(sa.lognormals)) isoPortionLines() plot(acomp(sa.lognormals)) isoProportionLines() plot(acomp(sa.lognormals),center=TRUE) isoProportionLines() plot(rcomp(sa.lognormals)) isoProportionLines()