cfrnow 0.2.1

Patch release: no behaviour changes since 0.2.0.

• distspec is now on CRAN, so it is dropped from Remotes and installed from
  CRAN like cfrnow's other dependencies.

cfrnow 0.2.0

• fit_cfr() and simulate_linelist() take a Weibull() onset-to-death (and
  onset-to-recovery) delay, alongside LogNormal() and Gamma().
• Delay parameterisation now uses distspec's exported natural_params() instead
  of an internal helper; cfrnow requires distspec (>= 0.1.0).
• New "Stratified and partially-pooled CFR" vignette: no-, complete- and
  partial-pooling fits, with per-group summary() output.
• pp_check_cfr() runs a posterior-predictive check. It draws replicate line
  lists from the posterior, replays the real-time truncation, and compares
  observed death counts (and recoveries, in a two-outcome fit) and onset-to-death
  delays against the replicates.
• summary() takes an ascertainment_ratio argument to correct the CFR when
  fatal and non-fatal cases enter the line list at different rates. Defaults to 1.
• fit_cfr() accepts intercept-free CFR formulas (e.g. cfr ~ 0 + group),
  fitting one logit-CFR per group with cfr_prior on those coefficients.
• summary() reports a CFR per group for a cfr ~ group fit, one cfr[<group>]
  row per group.

cfrnow 0.1.0

First release.

• fit_cfr() estimates a real-time case fatality ratio from line-list data with a
  Bayesian mixture-cure survival model. It is registered as an epidist model
  type, so the CFR and the onset-to-death delay both take brms formulas.
• prepare_cfr_data() turns a line list into model inputs. It sorts each case, at
  a chosen observation cut-off, into an observed death, a resolved non-death, or a
  right-censored survivor.
• The onset-to-death delay (LogNormal or Gamma) can be co-estimated or held fixed.
  Hold it fixed and you get the Ghani/Nishiura estimator.
• Pass a recovery_date column and a two-outcome fit also times recoveries.
• Put a brms formula on the CFR or the delay for covariates or a time-varying
  CFR.
• simulate_linelist() builds line lists for testing and examples.
• summary() and print() report the corrected CFR, the delay moments,
  convergence diagnostics, and a flag for when the CFR is only weakly identified.