NEWS file for the mlt package

Changes in Version 1.6-1 (2024-11-18)

Bugfixes

• Try harder to obtain starting values meeting constraints, issue a warning if this failed.

Changes in Version 1.6-0 (2024-09-23)

Bugfixes

• Allow interval censoring using -Inf and Inf. Starting values are affected but the corresponding observations contribute zero to the log-likelihood.

• Make sure R() does not drop any unused levels (unlike factor()).

New features

• mmlt moved from tram allowing arbitrary mixes of continuous and discrete data, including exact and censored observations for the same variable and thus also allowing non-informative missings encoded as -Inf, Inf, see below. Corresponding logLik and estfun methods feature newdata and weights arguments, as the methods for mlt always did.

• R() translates missing response values to -Inf and Inf interval censoring. This means that missing values are allowed in response variables and the corresponding observations do not contribute to the log-likelihood. In contrast to na.omit, the (technical) dimension of the training data is not changed.

Changes in Version 1.5-2 (2024-08-24)

Bugfixes

• Handle Surv objects with smaller than zero values in R.Surv(..., as.R.interval = TRUE).

New features

• Add maxeval = 1000L for nloptr.

• Report nloptr more detailed status messages.

Changes in Version 1.5-1 (2024-04-11)

Bugfixes

• Remove warning when dealing with binary factors.

Changes in Version 1.5-0 (2023-12-13)

Bugfixes

• Do no use zero as lower bound (when running R(Surv(...), as.R.interval = TRUE)) except when zero is first death time.

New features

• Add slots to mmlt for externally computing transformations and derivatives thereof, needed by tram::mmlt.

Changes in Version 1.4-9 (2023-08-21)

Bugfixes

• Make sure R(Surv(...), as.R.ordered = TRUE) matches Kaplan-Meier for right-censored and possibly tied observations. Such responses give NPML estimates.

New features

• R(Surv(...), as.R.numeric = TRUE) can be used to fit smooth models by maximising the nonparametric likelihood.

Changes in Version 1.4-8 (2023-06-27)

New features

• Better support for frailty parameter.

Changes in Version 1.4-7 (2023-05-22)

Bugfixes

• Starting values might end up being NA; set to zero.

New features

• mlt(..., theta = theta, dofit = TRUE) returns a "fitted" mlt model with coefficients theta. If theta is missing, an unfitted model is returned.

Changes in Version 1.4-6 (2023-04-14)

Bugfixes

• Register internal methods.

New features

• Re-add support for nloptr in mltoptim().

Changes in Version 1.4-5 (2023-03-09)

Bugfixes

• Fix S3 argument mismatches.

• Try to detect negative standard errors in case optimisation failed silently.

Changes in Version 1.4-4 (2022-11-02)

New features

• Improve numerical stability in simulate.ctm.

Changes in Version 1.4-3 (2022-10-31)

New features

• predict and plot have new types involving logs.

Bugfixes

• plot() ignored lwd.

Changes in Version 1.4-2 (2022-06-30)

New features

• as.R.interval allows evaluation of nonparametric likelihood for smoothly defined models.

• New Laplace link function, contributed by Ainesh Sewak.

• New cauchit link function.

Bugfixes

• drop = TRUE in score for shift-scale models.

• Eliminate fallout of fix of PR#17616 in Rout.save files.

Changes in Version 1.4-1 (2022-03-31)

New features

• update allows fixing parameters.

Bugfixes

• Fix discrete gradient for shift-scale models.

Changes in Version 1.4-0 (2022-01-14)

New features

• Add infrastructure for shift-scale transformation models.

Documentation

• Gradient and estfun always returned negative scores, this is documented now.

Bugfixes

• Arguments passed to mlt via dots were silently ignored.

• Remove nloptr dependency for the time being.

Changes in Version 1.3-2 (2021-08-24)

New features

• Add as.double for response objects (replaces trtf:::.R2vec).

Changes in Version 1.3-1 (2021-03-22)

New features

• Improved numerical stability for censored data.

Changes in Version 1.3-0 (2021-03-03)

New features

• New argument as.R.ordered allowing numeric and survival responses to be coded as ordered factors, for nonparametric maximum likelihood estimation.

• Allow sparse model matrices. This is useful for nonparametric maximum likelihood estimation with many distinct outcomes.

• Some speed-ups.

Bugfixes

• Plotting of quantiles sometimes failed because inversion of cdf was not possible for certain quantiles. These are now removed before plotting.

• Fitting models to interval censored responses containing intervals c(-Inf, Inf) failed.

• Always return names score matrices and residuals.

Changes in Version 1.2-3 (2021-02-22)

New features

• Improve documentation.

Changes in Version 1.2-2 (2021-02-12)

Bugfixes

• Sampling from unconditional models did not pay attention to number of observations.

Changes in Version 1.2-1 (2021-02-03)

New features

• Quantiles and thus simulations are now computationally more exact and more robust. The unnecessary interpolate argument to predict and simulate is now ignored.

• Adjust contrasts a fixed parameter contributes to.

• Return numerically determined Hessians upon request.

• Implement frailty error distributions, experimentally and internal only.

• Implement cure mixture models, experimentally and internal only.

• Improve computations of log-probabilities.

Bugfixes

• Discrete hazard functions were incorrect.

Changes in Version 1.2-0 (2020-05-11)

• Add exponential distribution (for Aalen additive hazards models).

• Pay attention to model class when computing cumulative hazards.

• Add log-cumulative hazards, log-odds, and odds for predictions and plots.

Changes in Version 1.1-2 (2020-02-13)

• Allow permutations of single variables.

• Update citation info.

Changes in Version 1.1-1 (2020-01-09)

• Try harder to invert Hessians.

• Update reference output.

Changes in Version 1.1-0 (2019-11-06)

• Add support for nloptr (still experimental and thus switched off by default).

• Make sure coef() always returns named argument.

• Fix problem in as.Surv reported by Balint Tamasi.

Changes in Version 1.0-7 (2019-10-07)

• Less paranoia in ‘bugfixes.R’.

Changes in Version 1.0-6 (2019-10-04)

• Return Hessian for fixed parameters if requested.

• Fix subsetting problem in R.numeric.

• Allow to update offsets.

Changes in Version 1.0-5 (2019-06-20)

• Add a bread method.

• Check response variable against observations in data.

• Make sure integers larger zero are handled correctly in R.

• Implement resid method, ie the score wrt a constant.

• Cox examples with Bernstein polynomials of log-time.

Changes in Version 1.0-4 (2018-12-03)

• Arguments K and cheat where ignored by confband when newdata had multiple rows.

• Computation of starting values more robust now.

• Order of fixed parameters (fixed argument to mlt) might have been wrong due to incomplete matching.

Changes in Version 1.0-3 (2018-09-17)

• Add lty argument to plot.ctm.

• update needs free coefficients only.

• Internal interface changes.

Changes in Version 1.0-2 (2018-06-22)

• Make sure transformation functions outside bounds are minus or plus Inf.

• Initial guestimates for ordered responses were incorrect and may potentially have led to nonsense results.

• Some smaller improvements in computation of log-likelihoods and scores with respect to accuracy and speed.

• print respects options(digits).

Changes in Version 1.0-1 (2018-04-20)

• estfun, parm = coef(object, fixed = TRUE)) evaluates scores for all model parameters, including fixed ones.

• logLik(..., newdata, w) ignored weights w when newdata was given. Same problem was also fixed for estfun.

Changes in Version 1.0-0 (2018-03-07)

• A paper describing version 1.0-0 of the mlt, basefun, and variables packages was accepted for publication in the Journal of Statistical Software 2018-03-05.

• Documentation updates.

Changes in Version 0.2-2 (2018-02-23)

• Use coneprog for getting the starting values.

• logLik and estfun accept matrices as parm argument for the evalution of log-likelihoods and scores with subject-specific parameters (for example in transformation trees or forests and boosting procedures.

Changes in Version 0.2-1 (2017-12-12)

• q is forwarded to qmlt by predict.ctm now.

• p is now prob in qmlt and thus predict.ctm.

• Update citation.

Changes in Version 0.2-0 (2017-06-20)

• Most Likely Transformations will be published in the Scandinavian Journal of Statistics.

• Import package alabama.

• as.Surv(R(Surv(...))) returns Surv(...), useful for converting output by simulate to Surv objects.

Changes in Version 0.1-3 (2017-04-04)

• Add subset argument to update (for faster transformation trees and forests).

• Sum over score contributions with positive weight only when evaluating the gradient.

• Having all response observations being interval-censored is allowed again (too heavy checking was in place).

• Don't try to numerically check KKT conditions automatically.

• Check for unused arguments in dots where necessary.

• Make sure the score doesn't get too large (avoid division by near zero probabilities).

• Improve survfit to compute non-parametric unconditional probabilities for obtaining starting values in the presence of censoring and truncation.

Changes in Version 0.1-2 (2017-02-14)

• logLik with newdata argument ignored parm and weights arguments.

• estfun now also has a newdata argument.

• Correct axes labelling when plotting quantile functions.

• make sure names are correct in coef(model, fixed = FALSE).

• check if any exact or interval-censored response with non-zero weight exists before trying to fit the model.

• make checks a little more robust against huge diffs under Windows.

Changes in Version 0.1-1 (2016-11-29)

• Fix two bugs in computation of log-likelihood for possibly left-truncated responses such as Surv(start, time, status).

Changes in Version 0.1-0 (2016-11-10)

• Add augmented lagrangian minimization (auglag() from package alabama).

• Make optimiation procedure more general and adaptive, allow users to change defaults and even add their own optimiser.

Changes in Version 0.0-34 (2016-09-13)

• fix bug when calling survfit for computing initial probabilities.

• add bysim argument to simulate.

• make sure checkGrad is respected by update.

• predict computes q with K elements if not given (as plot always did).

Changes in Version 0.0-33 (2016-05-25)

• Make sure times are ordered before calling survival::summary.survfit.

Changes in Version 0.0-32 (2016-05-24)

• Introduce as.mlt generic.

• Introduce a coef slot in ctm objects and a corresponding coef<- and coef method for setting and extracting coefficients to and from unfitted conditional transformation models.

• predict, simulate and plot work on ctm objects (with meaningful coefficients) now.

Changes in Version 0.0-31 (2016-03-11)

• Some small improvements wrt run time and memory consumption.

• Use theta = coef(object) as default starting parameters in update().

• logLik has a new newdata argument.

• simulate has a new q argument.

Changes in Version 0.0-30 (2016-02-19)

• The mlt package was first published on CRAN.