How to calculate scatter index in R statistics

Question

I am trying to figure out how to calculate the scatter index of a scatter plot in R. While I have the formula, I am still very new to using R and coding and am unsure of how to write it in the programming language. The formula is:

My data is:

structure(list(S = c(0.619, 0.582, 0.653, 0.611, 0.649, 0.558 ), 
               O = c(0.58, 0.55, 0.52, 0.56, 0.49, 0.54)), 
          row.names = c(NA, 6L), class = "data.frame")

Any help would be greatly appreciated.

What do observed and simulated values have to do with scatterplots? Can you link to a resource explaining this? — Joseph Clark McIntyre, Jan 08 '19 at 06:01
Here is a link to a paper that uses the above formula to calculate SI. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.724.9537&rep=rep1&type=pdf — R.Rookie, Jan 08 '19 at 06:03
Can you post the head of your data (run dput(head(data)) and paste the result into your question)? More importantly, does your dataset have a column for observed and simulated? If so, this should be very simple. — Joseph Clark McIntyre, Jan 08 '19 at 06:10
structure(list(S = c(0.619, 0.582, 0.653, 0.611, 0.649, 0.558 ), O = c(0.58, 0.55, 0.52, 0.56, 0.49, 0.54)), row.names = c(NA, 6L), class = "data.frame") — R.Rookie, Jan 08 '19 at 06:16

r2evans · Answer 1 · 2019-01-08T06:39:35.180

2

Here's a stab at your formula:

ScatterIndex <- function(s, o) sqrt( sum((s-mean(s)) - (o-mean(o)))^2 / sum(o^2) )

The vectors provided must be the same length.

Your data, and the output:

x <- structure(list(S = c(0.619, 0.582, 0.653, 0.611, 0.649, 0.558 ), O = c(0.58, 0.55, 0.52, 0.56, 0.49, 0.54)), row.names = c(NA, 6L), class = "data.frame")
ScatterIndex(x$S, x$O)
# [1] 1.676298e-16

edited Jan 08 '19 at 06:39

answered Jan 08 '19 at 06:13

r2evans

141,215
6
77
149

How to calculate scatter index in R statistics

1 Answers1