The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
The goal of this vignette is to help users understand the overall structure of the rsmart package. While the ‘Reference’ page on the pkgdown website is helpful for examining functions individually, we want to provide a high-level view of the package’s conceptual structure.
The rsmart package has 3 user-facing functions, along with a hierarchy of internal helpers. These functions can be organized into five conceptual groups, each of which we will unpack here in this vignette.
iaipwe() — The main estimation
workhorse. Orchestrates the entire pipeline: computes kappa, nu,
propensity scores, regime values, and then the sandwich variance (or
bootstrap variance). This is what users call directly.gen_no_trt_resp() — Data generation
for simulation studies.regime_list_no_trt_resp() — Builds the
regime/indicator matrices that iaipwe() needs.iaipwe)These are called first inside iaipwe() to
estimate the building blocks:
get_kappa() — Computes how far each
individual progressed (stage reached).get_nu() — Estimates stage-arrival
probabilities \(\nu_k\).pi_fits() — Fits propensity models at
all stages (calls pstep() per stage).iaipwe)estimate_values() — Loops over regimes
and for each one:
get_q_fits() — Fits Q-functions
backwards through stages (calls qstep()
per stage).value_terms() — Computes the \(2K+1\) coarsening-level value terms
(augmentation + IPW).get_bn() assembles \(\Psi_i \Psi_i^T / n\) by collecting
individual-level estimating equation contributions:
ee_psi_pi() — Estimating equation
contributions for \(\pi\)
parameters.ee_psi_nu() — Estimating equation
contributions for \(\nu\)
parameters.ee_psi_beta() — Estimating equation
contributions for \(\beta\)
(Q-function) parameters.ee_psi_v() — Estimating equation
contributions for the value parameters \(V\).get_an() assembles \(-\partial\Psi/\partial\theta\) by
collecting derivatives:
ee_dpsi_pi() — Derivative block for
\(\pi\).ee_dpsi_nu() — Derivative block for
\(\nu\).ee_dpsi_beta() — Derivative block for
\(\beta\).ee_dpsiv() (AIPW) or
ee_dpsiv_ipw() (IPW) — Derivative rows for
\(V\), which internally call:
ee_dpsiv_dpi() /
ee_dpsiv_dpi_ipw()ee_dpsiv_dnu() /
ee_dpsiv_dnu_ipw()ee_dpsiv_dbeta() (AIPW only)ee_dpsiv_dv()get_bounds() — Group sequential
stopping boundaries (calls
get_first_bound() +
get_next_bound()).get_sample_size() — Sample size
determination.get_q_coefs() — Coefficient extraction
utility.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.