[R] 3D scatterplot and agreement among three methods in the same subject

Mario Petretta petretta at unina.it
Fri Oct 31 16:31:00 CET 2014


Dear all,

I use R 3.1.1 for Windows.

I would like to use three-dimensional scatterplots to evaluate the agreement
among 3 laboratory methods in a set of 30 subjects (all subjects was
evaluated with each of the three methods), using a line of best fit (slope
and intercept) and the line of identity.

Moreover, I would like to evaluate the SD of the residual errors (distance
from line of best fit) to assess the disagreement between any method and the
average value among all 3  methods. For this purpose, no single method was
used as a standard to which the other 2 methods would be compared. 

The file is:

Id 	method1	method2	meth3		
1	5		6		6	    	
2	4		7		6
3	8		9		12
..	...		...		...
29	8		7		10
30	12		14		13

I know that the package "scatterplot3d" is useful for plotting the data, and
also I read the previous post
http://r.789695.n4.nabble.com/Fit-a-3-Dimensional-Line-to-Data-Points-td8635
96.html . However the R square and SD of the residual errors are not
reported in this example.

 
Any suggestion is welcome.

-------------------------------------------------------
Mario Petretta
Associate Professor of Internal Medicine
Department of Translational Medical Sciences
Naples University Federico II Italy



More information about the R-help mailing list