Compare the total log-likelihood values for the patient-level dataset after fitting PSM-simple and PSM-complex models to each combination of endpoint distributions
Arguments
- ptdata
Dataset of patient level data. Must be a tibble with columns named:
ptid: patient identifier
pfs.durn: duration of PFS from baseline
pfs.flag: event flag for PFS (=1 if progression or death occurred, 0 for censoring)
os.durn: duration of OS from baseline
os.flag: event flag for OS (=1 if death occurred, 0 for censoring)
ttp.durn: duration of TTP from baseline (usually should be equal to pfs.durn)
ttp.flag: event flag for TTP (=1 if progression occurred, 0 for censoring).
- fitslist
List of distribution fits to relevant endpoints, after calling
fit_ends_mods_par()orfit_ends_mods_spl()- cuttime
Time cutoff - this is nonzero for two-piece models.
Value
List containing
results: Dataset of calculation results for each modelbests: Tibble indicating which is the best fitting model individually or jointly, to each endpoint, according to AIC or BIC
Examples
# Fit parametric distributions to a dataset
bosonc <- create_dummydata("flexbosms")
parfits <- fit_ends_mods_par(bosonc)
# \donttest{
splfits <- fit_ends_mods_spl(bosonc)
# Present comparison of likelihood calculations
compare_psm_likes(bosonc, parfits)
#> Calculating PSM simple
#> Calculating PSM complex
#> Wrapping up
#> $results
#> # A tibble: 252 × 13
#> id ttp_meth pfs_dist os_dist ll npar npts aic bic rank_aic
#> <int> <chr> <chr> <chr> <dbl> <dbl> <int> <dbl> <dbl> <dbl>
#> 1 1 simple exp exp -859. 3 204 1725. 1735. 139
#> 2 2 simple exp weibullPH -834. 4 203 1676. 1689. 38
#> 3 3 simple exp llogis -839. 4 203 1687. 1700. 93
#> 4 4 simple exp lnorm -839. 4 203 1686. 1699. 91
#> 5 5 simple exp gamma -835. 4 203 1677. 1690. 42
#> 6 6 simple exp gompertz -830. 4 203 1668. 1682. 24
#> 7 7 simple weibullPH exp -851. 4 204 1711. 1724. 126
#> 8 8 simple weibullPH weibullPH -835. 5 203 1680. 1696. 53
#> 9 9 simple weibullPH llogis -841. 5 203 1692. 1709. 108
#> 10 10 simple weibullPH lnorm -839. 5 203 1687. 1704. 95
#> # ℹ 242 more rows
#> # ℹ 3 more variables: rank_bic <dbl>, best_aic <dbl>, best_bic <dbl>
#>
#> $bests
#> # A tibble: 4 × 5
#> ttp_meth pfs_dist os_dist meth ic
#> <chr> <chr> <chr> <chr> <chr>
#> 1 gompertz exp weibullPH ind aic
#> 2 gompertz exp weibullPH ind bic
#> 3 gompertz lnorm gamma joint aic
#> 4 gompertz lnorm gamma joint bic
#>
compare_psm_likes(bosonc, splfits)
#> Calculating PSM simple
#> Calculating PSM complex
#> Wrapping up
#> $results
#> # A tibble: 810 × 16
#> id ttp_meth ttp_knots pfs_scales pfs_knots os_scales os_knots ll npar
#> <int> <chr> <dbl> <chr> <int> <chr> <int> <dbl> <dbl>
#> 1 1 simple 0 hazard 3 hazard 3 -835. 7
#> 2 2 simple 0 hazard 3 odds 3 -840. 7
#> 3 3 simple 0 hazard 3 normal 3 -838. 7
#> 4 4 simple 0 hazard 3 hazard 4 -833. 8
#> 5 5 simple 0 hazard 3 odds 4 -833. 8
#> 6 6 simple 0 hazard 3 normal 4 -832. 8
#> 7 7 simple 0 hazard 3 hazard 5 -833. 9
#> 8 8 simple 0 hazard 3 odds 5 -832. 9
#> 9 9 simple 0 hazard 3 normal 5 -832. 9
#> 10 10 simple 0 odds 3 hazard 3 -825. 7
#> # ℹ 800 more rows
#> # ℹ 7 more variables: npts <int>, aic <dbl>, bic <dbl>, rank_aic <dbl>,
#> # rank_bic <dbl>, best_aic <dbl>, best_bic <dbl>
#>
#> $bests
#> # A tibble: 4 × 8
#> ttp_meth ttp_knots pfs_scales pfs_knots os_scales os_knots meth ic
#> <chr> <dbl> <chr> <int> <chr> <int> <chr> <chr>
#> 1 normal 3 normal 3 hazard 3 ind aic
#> 2 normal 3 normal 3 hazard 3 ind bic
#> 3 odds 3 normal 4 hazard 4 joint aic
#> 4 simple 0 odds 3 normal 4 joint bic
#>
# }