Output Columns ([output])
Maturity: beta — see Feature Maturity for what this means.
The optional [output] block lists additional columns to include in the sdtab beyond the mandatory minimum. Each entry is a bare name, one or more per line (whitespace or newlines between names).
[output]
WT AGE SEX
CLiWhat can be listed
| Category | Example names | Notes |
|---|---|---|
| Covariates | WT, AGE, SEX |
Written as-is from the data |
| Individual parameters | CL, V |
EBE-derived per-subject value |
| Derived column names | AUC, Cmax |
Defined in the [derived] block |
| Eta names | ETA_CL, ETA_V |
ETAs are not sdtab columns; declaration is ignored with W_OUTPUT_DUPLICATE |
Mandatory minimum (always present)
The following columns are always written to sdtab regardless of [output]:
ID, TIME, DV, CENS, OCC, CMT, PRED, IPRED, CWRES, IWRES, EBE_OFV, N_OBS, TAFD, TAD
Listing any of these in [output] is harmless but will emit W_OUTPUT_DUPLICATE.
CWRES is the conditional weighted residual of Hooker et al. (2007) as NONMEM tables it: the FOCE-linearized residual y − f0 (f0 = IPRED − H·η̂) decorrelated by the symmetric inverse square root of its covariance HΩHᵀ + R, with R evaluated at IPRED under an interaction fit and at the population prediction f(η = 0) under FOCE. Rebuilt from NONMEM 7.5.1’s own tabled G, ETA, IPRED and PRED, that recipe reproduces its CWRES column to 1e-4 (#1182). IWRES is (y − IPRED) / SD(IPRED), the individual weighted residual. Both are NaN on censored records.
Diagnostics
| Code | Severity | Cause |
|---|---|---|
E_OUTPUT_UNKNOWN_COLUMN |
error | A name is not a covariate, individual parameter, or derived expression |
W_OUTPUT_DUPLICATE |
warning | A name is already in the mandatory minimum |
Interaction with [derived]
A name defined in [derived] is automatically available in [output] without needing to be listed again — it is written to sdtab as soon as it appears in [derived]. Use [output] for covariates and individual parameters that are not computed by [derived].