The parameter key used in the pmparams functions can either be a yaml or data frame. It must contain the following:
name or yaml variable= THETA, OMEGA, SIGMA. See examples below.
abb = abbreviation you want to appear in the parameter table (use latex coding)
desc = description you want to appear in the parameter table
panel = indicate which panel each parameter should appear under.
Current options include:
panel=="struct" ~ "Structural model parameters"
panel=="cov" ~ "Covariate effect parameters"
panel=="IIV" ~ "Interindividual covariance parameters" (off-diagonals << function takes care of this)
panel=="IIV" ~ "Interindividual variance parameters" (diagonals << function takes care of this)
panel=="IOV" ~ "Interoccasion variance parameters"
panel=="RV" ~ "Residual variance"
trans = define how the parameter will be transformed. Current options include:
"none" - untransformed parameters, e.g. THETAs or off-diagonals
"logTrans" - THETAs estimated in the log domain
"logitTrans" - THETAs estimated using a logit transform
"lognormalOm"- for log-normal omegas e.g. CL = THETA(1) * exp(ETA(1)) - returns est.+CV%
"OmSD" - for omegas where you want to return SD only - returns estimate & SD; use this for additive omegas e.g. CL = THETA(1) + ETA(1)
"logitOmSD" - for omegas using logit transform - returns estimate & SD (calculated with logitnorm package); this option requires you provide the associated THETA separated with a "~"; e.g. "logitOmSD ~ THETA3"
"addErr" - for additive error terms (coded using SIGMA in $ERROR) - returns est.+ SD
"propErr" - for proportional error terms (coded using SIGMA in $ERROR) - returns est.+CV%
"addErrLogDV" - for additive error log terms (coded using SIGMA in $ERROR) - returns est.+CV%
Example of YAML syntax for parameter key file
THETA1: "KA (1/h)"
abb: "First order absorption rate constant"
desc: struct
panel: logTrans
trans
THETA2: "V2/F (L)"
abb: "Apparent central volume"
desc: struct
panel: logTrans trans
# Example of parameter key data frame:
# It must contain parameter names, abb, desc, panel, and trans columns.
paramKey = dplyr::tribble(
~name, ~abb, ~desc, ~panel, ~trans,
"THETA1", "KA (1/h)", "First order absorption rate constant", "struct", "logTrans",
"THETA2", "V2/F (L)", "Apparent central volume", "struct", "logTrans"
)
paramKey
#> # A tibble: 2 × 5
#> name abb desc panel trans
#> <chr> <chr> <chr> <chr> <chr>
#> 1 THETA1 KA (1/h) First order absorption rate constant struct logTrans
#> 2 THETA2 V2/F (L) Apparent central volume struct logTrans