covnmr_method_validate
ICH Q2(R2) method validation: linearity, precision, LOD/LOQ.
`concentrations` and `responses` are comma-separated series of equal
length, at least three points ("1,2,5,10,20"). Linearity is tested with
Mandel's F-test -- a linear and a quadratic fit compared against
F(0.05) -- so `linearity_confirmed` is a statistical verdict on curvature,
not an R-squared threshold.
Optional: `precision_measurements` (comma-separated repeat determinations)
adds the USP RSD check; `noise_rms` (> 0) adds LOD/LOQ from the calibration
slope. Both keys are ABSENT from the response when their input was not
given, and `overall_passes` is the AND of whichever checks actually ran --
so a run without precision data cannot pass on a criterion it never
evaluated.
Bounded: `max_points` (1000) per input series, `max_residuals` (500) on the
residual vector echoed back.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| api_key | No | ||
| noise_rms | No | ||
| responses | Yes | ||
| max_points | No | ||
| max_residuals | No | ||
| concentrations | Yes | ||
| precision_measurements | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |