Menu
Design a compact parameter-recovery simulation plan for the freqTLS 4PL thermal-load-sensitivity model.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Audit the freqTLS-versus-bayesTLS benchmark for a fair configuration, sound cache provenance, and the correct R-SHRIMP data rebuild.
Audit and improve freqTLS figures, galleries, pkgdown articles, and ggplot recipes, enforcing the Confidence-Eye uncertainty contract, before Florence, Rose, Pat, Fisher, and Grace call a figure done.
Review freqTLS before a public, CRAN, GitHub, or internal release.
Add simulation-based parameter-recovery tests for freqTLS models.
Audit a completed freqTLS task or phase before closing it, checking implementation, equations, examples, tests, docs, pkgdown, roadmap, NEWS, the capability matrix, known limitations, stale wording, and after-task reporting.
Review freqTLS profile-likelihood confidence intervals for correctness, equivariance, chi-square calibration, and honest open/boundary/multimodal handling.
| name | simulation-test-plan |
| description | Design a compact parameter-recovery simulation plan for the freqTLS 4PL thermal-load-sensitivity model. |
Use this skill when planning the simulation evidence for a freqTLS slice (recovery, coverage, edge behaviour, or benchmark sanity). Curie leads; Fisher reviews the inferential targets.
CTmax, z, low, up, k,
phi; profile equivariance ci_z == exp(ci_log_z); |D(MLE)| ~ 0; or a
warning on a weakly identified design.simulate_tls() (locked
data-generating process, fixed seed, factorial temperature x duration grid).fit_tls() workflow and form intervals with
confint(method = "profile").CTmax, degrees-per-decade for z, probabilities for low/up).phi near the binomial limit).data-raw/.CTmax to about 0.4 deg C, z
to about 0.6, low/up to about 0.05, k to about 30% relative; wider for
beta-binomial.logLik(beta_binomial) > logLik(binomial) and AIC(beta_binomial) < AIC(binomial) on overdispersed data; near-binomial on clean data.CTmax ~ 1 deg C, z ~ 25%); no Stan in the test.Too few temperatures or durations; no mortality; all mortality; threshold never
crossed; asymptote not approached; CTmax extrapolated; non-closing profile
(warning + NA endpoint, no crash); grouped designs with shared shape.