Gnuplot summing y values for same x values

Question

I have a dataset which looks like this:

0 1 0.1
0 0 0.1
0 1 0.1
1 0 0.2
0 1 0.2
1 0 0.2
...

I now want to do the following operations on each different value in the third column of the table:

Example for 0.1:

First column values summed: 0+0+0=0

Second column values summed: 1+0+1=2

Now I want to substract these two 2-0=2 and in a last step divide them by the occurrences.

2/3 =0.667

The same for 0.2 and my plot should then plot at x=0.1, y=0.667.

I hope my problem is with the example understandable.

Answer 1

You can use the smooth unique option to do exactly this: sum up all y-values belonging to the same x-value and then divide the result by the number of occurences. For the second column, upon which the operation is performed, you use the difference between the second and first column:

plot 'file.txt' using 3:($2 - $1) smooth unique

However, it seems like you'll run in a strange bug then. This works only correct, if you insert an empty or commented row at the beginning of your data file:

The result with the following file.txt

#
0 1 0.1
0 0 0.1
0 1 0.1
1 0 0.2
0 1 0.2
1 0 0.2

is