Project SOC-only curves under proportional hazards scenarios
Source:R/soc_projection.R
project_soc_only.RdProject SOC-only curves under proportional hazards scenarios
Usage
project_soc_only(
data,
time = "time",
event = "event",
subgroup = NULL,
arm = NULL,
soc_level = "SOC",
tau = NULL,
landmark_times = NULL,
hr_scenarios = c(0.7, 0.85, 1),
target_delta_rmst = NULL,
prior_mean_log_hr = NULL,
prior_sd_log_hr = NULL,
prior_draws = 2000L,
bootstrap = 0L,
seed = 1L,
n_grid = 100L
)Arguments
- data
Analysis-ready data frame.
- time, event
Column names identifying follow-up and event status.
- subgroup
Optional subgroup column name.
- arm
Optional treatment column. When supplied, only rows matching
soc_levelare retained before projection.- soc_level
Reference level used when
armis supplied.- tau
RMST truncation horizon.
- landmark_times
Survival-difference time points.
- hr_scenarios
Numeric vector of proportional-hazard multipliers.
- target_delta_rmst
Optional target RMST gain used to reverse-solve
required_hr.- prior_mean_log_hr, prior_sd_log_hr
Optional normal prior parameters for the log hazard ratio. When both are supplied alongside
target_delta_rmst, a PoS proxy is reported.- prior_draws
Number of Monte Carlo draws for the PoS proxy.
- bootstrap
Number of bootstrap resamples used to derive intervals.
- seed
Random seed.
- n_grid
Number of time points used to summarize curves.