Corrected the package citation to use the installed package version.
Standardized DOI markup in the reference manual for CRAN checks.
Changed the fit_mm() default from a log-only fit to
variance = "screen". It compares the four fixed shapes
constant, log, sqrt and
cuberoot, selects the lowest quasi-AIC by default, and
returns the selected fit_mm object with
screen_table (also variance_candidate_table).
Printing the fit or its summary includes the comparison and selection
indicator. Set variance = "log" to preserve the former
default. Explicit fixed shapes and user functions still fit directly;
variance = "auto" retains its separate log/power-grid
search. The existing power_selection argument also controls
the four-shape screen. Its intervals and bootstrap refits condition on
the selected shape and omit model-selection uncertainty.
Added method-specific references to the package overview, single-curve profile fitting help. These identify the conditional-moment framework, separable least-squares computation and wild-bootstrap multipliers.
Added full references and original-data locations for both bundled datasets in the data help pages, generated PDF reference manual and workflow guide.
Restored the 3.90625 micromolar reactions in
sdl_demo: six enzymes now have eight concentrations and
four reaction-level rates per concentration (192 rows, 32 per enzyme).
The previous 168-row example omitted unit-free lowest-concentration
headers during data preparation.
Added reproducible source-fluorescence preparation, a header parser that preserves decimals and accepts unit-free headers, and layout/rate regression checks. Both the R data and CSV mirrors are updated.
Clarified that group_mm() fits separate curves and
does not adjust residual clustering, that its model-based intervals
assume conditional independence, and that the demo’s replicate ordinal
is not a confirmed sample/plate/run ID.
Updated examples and data provenance. The fixed-shape estimating equations and numerical algorithms are retained; the new default selection and printing behavior changes the interface. Analyses that require a particular shape should specify it explicitly.
cluster_mm(method = "projected"), an
MM-specific Ma-Genton-style conditional projected-score construction
with the identity instrument. It retains conditionally Gaussian
residuals but does not specify the random-effect density. A scalar Km
score is followed by explicit moment updates and a stacked
independent-cluster sandwich covariance, not the legacy mean-information
approximation or a general efficiency guarantee. Refits retain this
method; pointwise mean-confidence intervals are available, but
distribution-free prediction quantiles are not. The legacy profile
method remains the default. Root, boundary and covariance checks stop
invalid fits without returning estimates or intervals.convergence; failures carry class
inferMM_profile_error, also inheriting from
inferMM_convergence_error. No estimates or intervals are
returned after a validation error. Automatic variance selection records
failed candidates and selects only successful fits. Screening/grouping
and the higher-moment profile pilot share the new root checks; the
separate clustered solver is unchanged.inferMM_convergence_error; no fit or confidence intervals
are returned. Non-Gaussian continuation failures suggest a manual retry
with normal working moments and the same variance function, without
automatic fallback or a guarantee of convergence or non-Gaussian
efficiency.fit_mm_moments() estimator for
independent curves. It jointly solves MM mean/variance estimating
equations weighted using third and fourth conditional moments, without
calling nls().fit_mm(), variance screening/grouping, and
clustered estimators retain their defaults. The new estimator does not
claim automatic finite- sample coverage, robustness to an incorrect
variance shape, or dominance over all weighted or transformation-based
methods.fit_mm() accepts a fixed user-defined positive function
of concentration; variance_function() can attach a
descriptive label to it.screen_mm() and group_mm() accept named
variance_functions lists for comparison with built-in
candidates, including custom-only screening.set.seed() outside
package functions; random-number handling is unchanged.nls fitting path so all single-curve fits
now use the manuscript’s Michaelis-Menten profile-score estimator.nls fallback.