[R] Results Differ in Ternary Plot Matrix of Compositional Response Variables

Rich Shepard rshepard at appl-ecosys.com
Fri May 1 20:32:32 CEST 2015

On Fri, 1 May 2015, David L Carlson wrote:

> Add plotMissings=FALSE to the second plot and see the plot.acomp manual
> page description of this argument:
> plot(JerrittY, pch=as.numeric(JerrittX4), col=c("black","red", "dark green",
>     "dark blue","dark goldenrod","dark orange","dark grey")[JerrittX4],
>      plotMissings=FALSE)


   Mea culpa! I overlooked the man page for plot.acomp.

   When I add plotMissings=FALSE I see no differences in the output despite
there being zeros for a comple of variables in two years.

   The horizontal bar with colors and labels on top of each ternary diagram
is what I want to eliminate and replace with the dot seen in plots with no
more than 6 rows (years) of data. There are 3 data sets with 6 rows, 1 with
7 rows (jerritt), and one with 8 rows and no missing data. The latter two
have the horizontal bars while the former three have dots.



More information about the R-help mailing list