Shift Analysis for Large Number of Tests
I’m trying to check for run-to-run variation across a large number of tests — e.g. >1,000 tests
The data is in long format, with all test values in a single column (TestValues
). Each unit has measurements across multiple runs (e.g., WW1, WW2, WW3), identified by a WW
column, and we have a UnitID
to track which unit it is. I want to:
Check if there’s a statistically significant shift across runs f...