We perform tests that measure performance of a device over time. The output is a value relative to the starting value. Right now to get to such relative performance, I setup a jmp sheet in such a way that I know what the reference is: Example:
Row Device Time Performance
1 T1 0 97
2 T2 0 93
3 T3 0 95
4 T1 100 83
5 T2 100 82
6 T3 100 83
For the relative performance, I setup a formula script
perf_0 = match (Device, "T1", Performance[1], "T2", Performance[2], "T3", Performance[3]);
Performance / perf_0
This works well with a limited number of data points, but is prone to typing errors for large data sets. Is there a better way to have this automated and less prone to errors? Note that the Performance value at time 0, is not necessarily the maximum or minimum value.