rSPDE 2.7.0

• rspde_safe_inla() now finds the INLA executable with INLA >= 26.11, which no longer sets the inla.call option. Previously it reported INLA as not installed correctly, so INLA examples and tests were skipped. A missing executable is now reported with a hint to run INLA::inla.stiles.install().
• The cross-validation routing tests skip the MetricGraph delegation checks when INLA is not available.
• Added type_rational_approximation = "wl2", a new way of obtaining the rational coefficients. The classes are parameterised so that every fit is a valid model and have no constant term, so the covariance-based models have m instead of m + 1 latent blocks. Available in matern.operators() and CBrSPDE.matern.operators() for both type = "covariance" and type = "operator", in matern.rational() and matern.rational.cov() for the exact one-dimensional models, and directly through rational.coefficients.wl2(). It requires floor(nu + d/2) to be 0, 1 or 2 for the covariance type, that is nu < 3 - d/2, and nu + d/2 < 2 for the operator type. The existing "brasil", "chebfun" and "chebfunLB" types are unchanged and remain the default.
• spde.matern.operators() accepts type_rational_approximation = "wl2" for both type = "covariance" and type = "operator". The coefficients depend on alpha and the dimension only, not on the varying kappa and tau, so the same ones serve the non-stationary model; the blocks carry the shift in the same way as the stationary ones.
• Fixed: the non-stationary covariance-based model (spde.matern.operators() with type = "covariance") joined its blocks outside the loop over the poles, so the precision had three blocks whatever the order. For m at least 3 the model was silently the wrong one, missing the poles 2 to m-1.
• Fixed: spde.matern.operators() with type = "operator" did not pass type_rational_approximation on, so every type gave the same model.
• type = "operator" now accepts only two of the four rational types. Its factorisation has a single table of roots, produced by the chebfun lower-bound method, so "chebfunLB" selects that table and "wl2" fits its own; "brasil" and "chebfun" are refused rather than quietly given roots they did not produce, and remain available for type = "covariance". Leaving type_rational_approximation at its default is unaffected. The tabulated roots are stored for m at most 4, while "wl2" fits them at any order. The operator-based models have no INLA interface; rspde.matern() is covariance-based throughout.
• rspde.matern1d() accepts type.rational.approx = "wl2", for a fixed nu and for an estimated one. The exact one-dimensional model then has rspde.order blocks of floor(alpha) + 1 entries per location instead of that plus a k-block, and the pole blocks carry the shared shift.
• rspde.matern() accepts type.rational.approx = "wl2", for a fixed nu and for an estimated one, with rspde.order at least 1. The latent field then has rspde.order blocks instead of rspde.order + 1, which rspde.make.A() and rspde.make.index() follow through their new type.rational.approx argument.
• The mesh-free weighted-L2 coefficients are stored in the package, as the tabulated ones are, for d = 1 to 3, m = 1 to 6 and floor(alpha) = 0, 1 and 2. They depend on neither the mesh nor kappa and are what a covariance-based model uses by default, so the common case fits nothing at set-up. data-raw/wl2_tables.R regenerates them, and the tests check that a fresh fit still reproduces what is stored.
• Added rspde.xmin(), which computes the lower end of the spectral interval from the mesh and a lower bound for kappa, and the arguments x_min and kappa_ref to matern.operators(). Fitting on that shorter interval is appreciably more accurate than the mesh-free fit, and is done when the model is created. update_rational_coefficients() recomputes the coefficients once kappa has been estimated.
• Added rspde.wl2.table(), which builds a weighted-L2 table for a given spectral interval, and the wl2_table argument of matern.operators(), which takes one and overrides x_min and kappa_ref. A table built for another dimension, order or range of alpha is refused rather than used.
• Added rspde.cache(), which keeps generated coefficient tables between sessions, under tools::R_user_dir("rSPDE", "cache") or a directory of your choosing. It is off by default, since a package should not write outside the session temporary directory unless asked; the environment variable RSPDE_CACHE_DIR sets it for non-interactive use. Lookup is automatic, and a table fitted on a slightly wider spectral interval is reused, since it still covers the whole spectrum.
• Added variance_correction = "nodal" to matern.operators(), which adds max(sigma^2 - diag(Sigma), 0) to the diagonal of the covariance of a "wl2" covariance-based model. What this corrects is mostly the finite element discretisation rather than the rational approximation. It is off by default, as it depends on the parameters it cannot be tabulated, so it is meant for a model whose parameters are already estimated.
• The package now has compiled code in every install, CRAN included: src/wl2_fit.cpp holds the inner loop of the weighted-L2 fit. The INLA cgeneric sources remain optional and are still built only with RSPDE_COMPILE=1 or --configure-args='--enable-compiled'. The equivalent R implementation is kept as the reference and is used when options(rSPDE.wl2.use.cpp = FALSE).
• The RSpectra dependency is gone. rspde.xmin(eigenvalue = "exact") and the scaling of the intrinsic models now use Lanczos iterations in the package. Both are also more robust, and the intrinsic one is faster than the old method.
• intrinsic.operators() now honours its opts argument, which was built and then replaced by a hardcoded list. Its entries are tol and maxitr, as before.
• Fixed matern.rational.cov(), which evaluated the covariance at the lags h[1] - h, rather than at h. A matrix of lags is now also accepted, and returns a matrix.
• get.roots() now uses spline interpolation by default. Linear interpolation lost accuracy off the 200-node beta grid of the tables (symbol error 1.6e-4 instead of 1.8e-6 at beta = 0.875, m = 4).
• Fixed the inlabru mapper for rspde.spacetime() models whose spatial mesh is a metric_graph. bru_get_mapper() used bm_fmesher() for the graph, which has no fm_dof() method, so bru() failed with “invalid subscript type ‘list’”.
• The optional cgeneric Makefile now only uses Homebrew gcc-14 on macOS and the compilers R was configured with elsewhere, so compiled installs (RSPDE_COMPILE=1) work on Linux.
• posterior_crossvalidation() is now an S3 generic, with methods for rspde_lme fits and for lists of fitted models. MetricGraph provides the graph_lme method, so the two packages no longer mask each other’s function, and a list can mix rspde_lme and graph_lme fits.

rSPDE 2.6.0

• Added posterior_crossvalidation() for objects fitted with rspde_lme(). The function mirrors the interface of MetricGraph::posterior_crossvalidation.
• Added hybrid.spde(), a new hybrid Whittle-Matern SPDE model with a non-zero deterministic mean.
• Added rspde.hybrid.matern(), a INLA cgeneric model for the hybrid Whittle-Matern SPDE with alpha = 2.
• Added a kappa_mu option that lets the operator applied to the mean use a different range parameter from the one in the covariance. Default (kappa_mu = NULL in hybrid.spde(), separate_kappa_mu = FALSE in rspde.hybrid.matern()) keeps them linked. When enabled, kappa_mu is estimated jointly in rspde_lme and INLA, or can be held fixed via model_options$fix_kappa_mu.
• The INLA cgeneric models now use the rSPDE models built into INLA when they are available, falling back to the local rSPDE shared library otherwise. shared_lib also accepts a path to a shared library file.
• Added rspde_safe_inla() and local_rspde_safe_inla(), which check that a usable INLA installation is available, for use in examples and tests.
• rspde.metric_graph() now passes shared_lib on to rspde.matern(), and its default is now "detect", matching the other INLA models.
• Updated the inlabru interface to the inlabru 2.14 API. rSPDE now requires fmesher (>= 0.7.0) and suggests inlabru (>= 2.14.0).
• predict.rspde_lme() can reuse precomputed parameter-dependent quantities, which speeds up repeated predictions such as in cross-validation.
• Fixed predict.rspde_lme() for models with replicates: the same location in different replicates no longer triggers a duplicated-locations warning.
• Fixed update() for non-stationary models, where new theta values were ignored, so predictions from non-stationary rspde_lme() fits used stale parameters.
• spde.matern.operators() no longer converts a model to a stationary one when B.tau or B.kappa vary in space.
• Added a vignette comparing rSPDE with the exact Matern covariance in terms of timing and memory.

rSPDE 2.5.2

• Added intrinsic Matern mapper support and related documentation.
• Added covariance_mesh, cov_function_mesh, and make_A documentation.
• Improved intrinsic and fractional operator implementations and stability.
• Updated INLA/inlabru interfaces and examples.
• Expanded unit tests and vignette updates (spacetime/anisotropic/intrinsic).

rSPDE 2.5.1

• Added model_options argument to rspde_lme() function, which allows users to set starting values for different parameters and also to fix parameters during estimation.
• Added previous_fit argument to rspde_lme(), which allows users to provide a previously fitted model as input to obtain starting values for a new fit.

rSPDE 2.5.0

• Improved the cross_validation function to allow for multiple likelihoods.
• General adjusts on rspde.intrinsic for stability.
• inlabru implementation for rspde.intrinsic.
• Improved warning messages when calling inla-related functions.
• Added wCRPS and swCRPS scores on cross_validation.

rSPDE 2.4.0

• Created the group_predict function, to obtain predictions on a testing set based on observations on a training set.
• Added support for stochvol, stochvol.nig, stochvolln and binomial likelihoods in cross_validation function.
• Changing the default nu.upper.bound to 2 in dimension 1, and keeping the default nu.upper.bound to 4 in dimension 2 in rspde.matern() function.
• Created matern.rational() operators for creating stationary matern operators.
• Created spacetime.operators() for creating space-time models.
• Created matern2d.operators() for anisotropic operators.
• Implemented space-time operators in cgeneric to be used in INLA and inlabru.
• Implemented anisotropic operators in cgeneric to be used in INLA and inlabru.
• Implemented stationary operators in cgeneric to be used in INLA and inlabru.
• Added vignette on space-time models.
• Added vignette on stationary models.
• Added vignette on anisotropic models.

rSPDE 2.3.3

• Bugfix on rspde_lme when fitting with fixed smoothness.
• Added a 2d fem interface.
• Moved from using INLA’s mesh functions to fmesher’s mesh functions.
• Removing rgdal from suggests.
• The data argument in predict.rspde_lme has been changed to newdata.
• Adding covariance_mesh and cov_function_mesh methods as functions in the list returned by objects obtained from matern.operators() and spde.matern.operators().
• Updated the internal structure to match the updates from the MetricGraph package.
• Updated the cross_validation function to match the updates in inlabru.
• Added glance and augment methods for rspde_lme objects.

rSPDE 2.3.2

• Small improvement on speed for rspde_lme.
• Bugfix on Q for small values of nu in dimension 1.
• Adding parameterization option for rspde.result.
• Bugfix on which_repl in rspde_lme.
• Addressing issues related to the new version of the Matrix package.

rSPDE 2.3.1

• Adding references in DESCRIPTION.
• Changing link to eigen library.

rSPDE 2.3.0

• Fixed a bug on rSPDE.construct.matern.loglike when the parameterization is “matern”.
• Created the rspde_lme() interface, with corresponding standard methods(predict, summary, etc).
• Updated the vignettes to use the rspde_lme() interface instead of the likelihood function factory.
• Replaced chol by Cholesky when using it to compute determinants or to solve systems.

rSPDE 2.2.0

• Adding a new parameterization (variance and a range-like parameter)
• Posterior sampling on the predict method.
• Added the cross_validation function which has several scoring rules implemented (MSE, CRPS, SCRPS, DSS) based on our inlabru implementation of the rational SPDE approach.

rSPDE 2.1.0

• Expanded the parameterization options on matern.operators and spde.matern.operators, along with their associated functions.
• Implementation of the precision method for inla_rspde objects.
• Implementation of the covariance-based spde.matern.operators function and its associated functions.
• Adjusts on the compatibility with the forthcoming MetricGraph package.

rSPDE 2.0.0

• Added cgeneric versions of the nonstationary models
• Added support for metric graphs (depends on the MetricGraph package)
• Added cgeneric versions of the stationary models
• Replaced rgeneric models by their cgeneric counterparts
• Added a new parameterization (range and std. dev)
• Created a new method gg_df to help posterior plotting in ggplot2

rSPDE 1.2.0

• Added an inlabru interface
• Added “rational.order” and “rational.type” functions
• Added the BRASIL rational approximation
• Improved covariance-based operator objects
• Improved log-likelihood computation
• Created 2d folded Matern under different boundary conditions
• Implemented different boundary conditions for 1d folded Matern

rSPDE 1.1.1

• Adjusts on donttest examples for CRAN

rSPDE 1.1.0

• Minor typos on vignettes and man pages were corrected
• Some examples were changed to improve their numerical stability

rSPDE 1.0.0

• Implementation of the covariance-based rational approximation for stationary Matérn models
• R-INLA implementation of the rational SPDE approach
• Added an introduction to rSPDE vignette
• The previous vignette was updated an became an operator-based rational approximation vignette
• Added a vignette for the R-INLA implementation of the SPDE approach
• Added a vignette to present the rational approximation using the rSPDE package
• Backward compatibility was maintained

rSPDE 0.6.3

• Change to inline citations in the Vignette to avoid problems on CRAN

rSPDE 0.6.2

rSPDE 0.6.1

• Add rgdal as suggested package

rSPDE 0.5.0

• Remove dependency on INLA for Vignette on CRAN
• Update citation