Generalized Fit Object (GFO)

Generalized fit object (GFO) stores population and individual estimation results, diagnostic tables, and variance–covariance matrices produced by model fitting.

Format

GFO

A named list with the following components:

SDTAB

Data frame. Standard diagnostic table of observations and model predictions. Typical columns: ID, TIME, DV (observed), DVID, DVNAME, PRED (population prediction), IPRED (individual prediction), RES, IRES, WRES, IWRES (residuals), MDV, and optionally OCC, BLQ, CENS, LIMIT.

SUMTAB

Data frame. Parameter summary table. Typical columns: PAR (parameter name), VALUE (estimate), TYPE ("Typical values", "Random effects", "Covariate coefficients", "Correlation coefficients", or "Residual error model"), DISTRIBUTION, EST ("ESTIMATED" or "FIXED"), SE, RSE, CV, LCI, UCI, ETAshrinkage_var, ETAshrinkage_sd, EPSshrinkage_sd.

SIGMAMAT

Numeric matrix. Residual error variance–covariance matrix (diagonal elements are residual variance parameters).

OMEGAMAT

Numeric matrix. Inter-individual variability variance–covariance matrix on the omega_* scale.

OCCMAT

Numeric matrix. Inter-occasion variability matrix; empty when occasion variability is not present.

EVTAB

Data frame. Dosing/event table extracted from the dataset. Typical columns: ID, TIME, EVID, CMT, ADM, AMT, and optionally OCC, ADDL, II, DUR, TINF, RATE, SS.

PATAB

Data frame. Individual random effects and post hoc parameter estimates. Columns ID, eta_* (individual ETAs), and individual PK/PD parameter columns (e.g. ka, Vd, CL).

COTAB

Data frame. Continuous covariates: ID plus continuous covariate columns (including derived columns such as log-transformed covariates). Empty data frame when none are present.

CATAB

Data frame. Categorical covariates: ID plus categorical covariate columns. Empty data frame when none are present.

REGTAB

Data frame. Time-varying regressors (ID, TIME, regressor columns). Empty data frame when none are present.

OFV

Data frame with one row. Model fit criteria parsed from Monolix summary.txt: LL (minus 2 times log-likelihood, Monolix OFV), AIC, BIC, BICc.

COVMAT

Numeric matrix. Variance–covariance matrix of population parameter estimates.

CORRMAT

Numeric matrix. Correlation matrix of population parameter estimates.

OPTIONS

List or NULL. Additional model or task options when available.

PROJNAME

Character. Project or run name.

Details

Create a GFO with sg_converter() from a Monolix project, or with sg_fit() when fit = TRUE.

See also

GCO, sg_fit(), sg_converter(), read_smrg_obj(), gfo4cov

Examples

# \donttest{
gfo <- read_smrg_obj(gfo4cov)
names(gfo)
#>  [1] "PROJNAME" "SUMTAB"   "SDTAB"    "PATAB"    "COVMAT"   "CORRMAT" 
#>  [7] "OFV"      "OMEGAMAT" "SIGMAMAT" "EVTAB"    "COTAB"    "CATAB"   
head(gfo$SDTAB)
#>   ID TIME     DV DVID   PRED  IPRED     RES    IRES WRES   IWRES MDV
#> 1  1 0.25 0.0098    1 0.0085 0.0090  0.0013  0.0007    0  1.4225   0
#> 2  1 0.50 0.0159    1 0.0169 0.0178 -0.0009 -0.0019    0 -1.8636   0
#> 3  1 1.00 0.0342    1 0.0330 0.0348  0.0012 -0.0005    0 -0.2713   0
#> 4  1 2.00 0.0611    1 0.0634 0.0661 -0.0023 -0.0050    0 -1.3087   0
#> 5  1 3.00 0.1028    1 0.0914 0.0944  0.0114  0.0084    0  1.5351   0
#> 6  1 6.00 0.1703    1 0.1621 0.1630  0.0082  0.0073    0  0.7735   0
head(gfo$SUMTAB)
#>                  PAR   VALUE              TYPE       EST     SE      RSE
#> 1             ka_pop  0.0671    Typical values ESTIMATED 0.0076  11.3325
#> 2      beta_ka_SEX_1 -0.1220 Covariate effects ESTIMATED 0.1307 107.1425
#> 3             Vd_pop 17.0503    Typical values ESTIMATED 0.9173   5.3802
#> 4  beta_Vd_LG_WEIGHT  0.6053 Covariate effects ESTIMATED 0.7682 126.9111
#> 5             CL_pop  0.2152    Typical values ESTIMATED 0.0382  17.7293
#> 6 beta_CL_CYP2C9_1_2 -0.3394 Covariate effects ESTIMATED 0.0801  23.5878
#>       LCI     UCI ETAshrinkage_var ETAshrinkage_sd EPSshrinkage_sd
#> 1  0.0522  0.0820               NA              NA              NA
#> 2 -0.3781  0.1342               NA              NA              NA
#> 3 15.2523 18.8483               NA              NA              NA
#> 4 -0.9003  2.1109               NA              NA              NA
#> 5  0.1404  0.2900               NA              NA              NA
#> 6 -0.4963 -0.1825               NA              NA              NA
head(gfo$PATAB)
#>   ID  eta_ka  eta_Vd  eta_CL     ka      Vd     CL
#> 1  1  0.3240  0.2633  0.1085 0.0821 22.4703 0.4220
#> 2  2 -0.1098  0.4202 -0.0021 0.0532 25.8580 0.3198
#> 3  3 -0.1721  0.2817 -0.7066 0.0500 23.8715 0.1302
#> 4  4 -1.0067 -0.5369  0.3402 0.0217  9.8194 0.5638
#> 5  5  0.0830  0.4890  0.0291 0.0645 28.2193 0.3787
#> 6  6 -0.1946  0.4149  0.1531 0.0489 26.1478 0.2166
gfo$OFV
#>          LL       AIC       BIC     BICc
#> 1 -10070.29 -10040.29 -10001.21 -9990.12
gfo$OMEGAMAT
#>        [,1]   [,2]   [,3]
#> [1,] 0.3053 0.0000 0.0000
#> [2,] 0.0000 0.2704 0.0000
#> [3,] 0.0000 0.0000 0.0846
# }