This package provides a collection of functions to deal with spatial and spatiotemporal autoregressive conditional heteroscedasticity (spatial ARCH and GARCH models) by Otto, Schmid, Garthoff (2018, Spatial Statistics) <doi:10.1016/j.spasta.2018.07.005>: simulation of spatial ARCH-type processes (spARCH, log/exponential-spARCH, complex-spARCH); quasi-maximum-likelihood estimation of the parameters of spARCH models and spatial autoregressive models with spARCH disturbances, diagnostic checks, visualizations.

This package provides a graph community detection algorithm that aims to be performant on large graphs and robust, returning consistent results across runs. SpeakEasy 2 (SE2), the underlying algorithm, is described in Chris Gaiteri, David R. Connell & Faraz A. Sultan et al. (2023) <doi:10.1186/s13059-023-03062-0>. The core algorithm is written in C', providing speed and keeping the memory requirements low. This implementation can take advantage of multiple computing cores without increasing memory usage. SE2 can detect community structure across scales, making it a good choice for biological data, which often has hierarchical structure. Graphs can be passed to the algorithm as adjacency matrices using base R matrices, the Matrix library, igraph graphs, or any data that can be coerced into a matrix.

Estimation of robust estimators for multi-group and spatial data including the casewise robust Spatially Smoothed Minimum Regularized Determinant (ssMRCD) estimator and its usage for local outlier detection as described in Puchhammer and Filzmoser (2023) <doi:10.1080/10618600.2023.2277875> as well as for sparse robust PCA for multi-source data described in Puchhammer, Wilms and Filzmoser (2024) <doi:10.48550/arXiv.2407.16299>. Moreover, a cellwise robust multi-group Gaussian mixture model (MG-GMM) is implemented as described in Puchhammer, Wilms and Filzmoser (2024) <doi:10.48550/arXiv.2504.02547>. Included are also complementary visualization and parameter tuning tools.

Semi-parametric estimation problem can be solved by two-step Newton-Raphson iteration. The implicit profiling method<arXiv:2108.07928> is an improved method of two-step NR iteration especially for the implicit-bundled type of the parametric part and non-parametric part. This package provides a function semislv() supporting the above two methods and numeric derivative approximation for unprovided Jacobian matrix.

Add a searchbar widget to your Shiny application. The widget quickly integrates with any existing element containing text to highlight matches. Highlighting is done with the JavaScript library mark.js'. The widget includes buttons to cycle through multiple instances of the match and automatically scroll to the matches in an overflow element (or window). The widget also displays the total number of matches and which match is currently being cycled through. The widget is structured as a Bootstrap 3 input group.

Set of functions to quantify and map the behaviour of winds generated by tropical storms and cyclones in space and time. It includes functions to compute and analyze fields such as the maximum sustained wind field, power dissipation index and duration of exposure to winds above a given threshold. It also includes functions to map the trajectories as well as characteristics of the storms.

Collection of datasets from Sen & Srivastava: "Regression Analysis, Theory, Methods and Applications", Springer. Sources for individual data files are more fully documented in the book.

Browser notifications in Shiny apps, using toastr': <https://github.com/CodeSeven/toastr#readme>.

The code computes the structural intervention distance (SID) between a true directed acyclic graph (DAG) and an estimated DAG. Definition and details about the implementation can be found in J. Peters and P. Bühlmann: "Structural intervention distance (SID) for evaluating causal graphs", Neural Computation 27, pages 771-799, 2015 <doi:10.1162/NECO_a_00708>.

The predictive value of a statistical model can often be improved by applying shrinkage methods. This can be achieved, e.g., by regularized regression or empirical Bayes approaches. Various types of shrinkage factors can also be estimated after a maximum likelihood. While global shrinkage modifies all regression coefficients by the same factor, parameterwise shrinkage factors differ between regression coefficients. With variables which are either highly correlated or associated with regard to contents, such as several columns of a design matrix describing a nonlinear effect, parameterwise shrinkage factors are not interpretable and a compromise between global and parameterwise shrinkage, termed joint shrinkage', is a useful extension. A computational shortcut to resampling-based shrinkage factor estimation based on DFBETA residuals can be applied. Global, parameterwise and joint shrinkage for models fitted by lm(), glm(), coxph(), or mfp() is available.

Computes the entire regularization path for the two-class svm classifier with essentially the same cost as a single SVM fit.

In stability selection (N Meinshausen, P Bühlmann (2010) <doi:10.1111/j.1467-9868.2010.00740.x>) and consensus clustering (S Monti et al (2003) <doi:10.1023/A:1023949509487>), resampling techniques are used to enhance the reliability of the results. In this package (B Bodinier et al (2025) <doi:10.18637/jss.v112.i05>), hyper-parameters are calibrated by maximising model stability, which is measured under the null hypothesis that all selection (or co-membership) probabilities are identical (B Bodinier et al (2023a) <doi:10.1093/jrsssc/qlad058> and B Bodinier et al (2023b) <doi:10.1093/bioinformatics/btad635>). Functions are readily implemented for the use of LASSO regression, sparse PCA, sparse (group) PLS or graphical LASSO in stability selection, and hierarchical clustering, partitioning around medoids, K means or Gaussian mixture models in consensus clustering.

The SPARRA risk score (Scottish Patients At Risk of admission and Re-Admission) estimates yearly risk of emergency hospital admission using electronic health records on a monthly basis for most of the Scottish population. This package implements a suite of functions used to analyse the behaviour and performance of the score, focusing particularly on differential performance over demographically-defined groups. It includes useful utility functions to plot receiver-operator-characteristic, precision-recall and calibration curves, draw stock human figures, estimate counterfactual quantities without the need to re-compute risk scores, to simulate a semi-realistic dataset. Our manuscript can be found at: <doi:10.1371/journal.pdig.0000675>.

Automatically calculate direct, indirect, and total effects for piecewise structural equation models, comprising lists of fitted models representing structured equations (Lefcheck, 2016 <doi:10/f8s8rb>). Confidence intervals are provided via bootstrapping.

Spatio-temporal change of support (STCOS) methods are designed for statistical inference on geographic and time domains which differ from those on which the data were observed. In particular, a parsimonious class of STCOS models supporting Gaussian outcomes was introduced by Bradley, Wikle, and Holan <doi:10.1002/sta4.94>. The stcos package contains tools which facilitate use of STCOS models.

This package provides a set of statistical tools for spatio-temporal data exploration. Includes simple plotting functions, covariance calculations and computations similar to principal component analysis for spatio-temporal data. Can use both dataframes and stars objects for all plots and computations. For more details refer Spatio-Temporal Statistics with R (Christopher K. Wikle, Andrew Zammit-Mangion, Noel Cressie, 2019, ISBN:9781138711136).

In clinical trials, endpoints are sometimes evaluated with uncertainty. Adjudication is commonly adopted to ensure the study integrity. We propose to use multiple imputation (MI) introduced by Robin (1987) <doi:10.1002/9780470316696> to incorporate these uncertainties if reasonable event probabilities were provided. The method has been applied to Cox Proportional Hazard (PH) model, Kaplan-Meier (KM) estimation and Log-rank test in this package. Moreover, weighted estimations discussed in Cook (2004) <doi:10.1016/S0197-2456(00)00053-2> were also implemented with weights calculated from event probabilities. In conclusion, this package can handle time-to-event analysis if events presented with uncertainty by different methods.

This package performs estimation and inference on a partially missing target outcome (e.g. gene expression in an inaccessible tissue) while borrowing information from a correlated surrogate outcome (e.g. gene expression in an accessible tissue). Rather than regarding the surrogate outcome as a proxy for the target outcome, this package jointly models the target and surrogate outcomes within a bivariate regression framework. Unobserved values of either outcome are treated as missing data. In contrast to imputation-based inference, no assumptions are required regarding the relationship between the target and surrogate outcomes. Estimation in the presence of bilateral outcome missingness is performed via an expectation conditional maximization either algorithm. In the case of unilateral target missingness, estimation is performed using an accelerated least squares procedure. A flexible association test is provided for evaluating hypotheses about the target regression parameters. For additional details, see: McCaw ZR, Gaynor SM, Sun R, Lin X: "Leveraging a surrogate outcome to improve inference on a partially missing target outcome" <doi:10.1111/biom.13629>.

This package performs repeated nested cross-validation for Cox Proportionate Hazards, Cox Lasso, Survival Random Forest, and their ensemble. Returns internally validated concordance index, time-dependent area under the curve, Brier score, calibration slope, and statistical testing of non-linear ensemble outperforming the baseline Cox model. In this, it helps researchers to quantify the gain of using a more complex survival model, or justify its redundancy. Equally, it shows the performance value of the non-linear and interaction terms, and may highlight the need of further feature transformation. Further details can be found in Shamsutdinova, Stamate, Roberts, & Stahl (2022) "Combining Cox Model and Tree-Based Algorithms to Boost Performance and Preserve Interpretability for Health Outcomes" <doi:10.1007/978-3-031-08337-2_15>, where the method is described as Ensemble 1.

This package provides a modification of the preventive vaccine efficacy trial design of Gilbert, Grove et al. (2011, Statistical Communications in Infectious Diseases) is implemented, with application generally to individual-randomized clinical trials with multiple active treatment groups and a shared control group, and a study endpoint that is a time-to-event endpoint subject to right-censoring. The design accounts for the issues that the efficacy of the treatment/vaccine groups may take time to accrue while the multiple treatment administrations/vaccinations are given; there is interest in assessing the durability of treatment efficacy over time; and group sequential monitoring of each treatment group for potential harm, non-efficacy/efficacy futility, and high efficacy is warranted. The design divides the trial into two stages of time periods, where each treatment is first evaluated for efficacy in the first stage of follow-up, and, if and only if it shows significant treatment efficacy in stage one, it is evaluated for longer-term durability of efficacy in stage two. The package produces plots and tables describing operating characteristics of a specified design including an unconditional power for intention-to-treat and per-protocol/as-treated analyses; trial duration; probabilities of the different possible trial monitoring outcomes (e.g., stopping early for non-efficacy); unconditional power for comparing treatment efficacies; and distributions of numbers of endpoint events occurring after the treatments/vaccinations are given, useful as input parameters for the design of studies of the association of biomarkers with a clinical outcome (surrogate endpoint problem). The code can be used for a single active treatment versus control design and for a single-stage design.

The explosion of biobank data offers immediate opportunities for gene-environment (GxE) interaction studies of complex diseases because of the large sample sizes and rich collection in genetic and non-genetic information. However, the extremely large sample size also introduces new computational challenges in GxE assessment, especially for set-based GxE variance component (VC) tests, a widely used strategy to boost overall GxE signals and to evaluate the joint GxE effect of multiple variants from a biologically meaningful unit (e.g., gene). We present SEAGLE', a Scalable Exact AlGorithm for Large-scale Set-based GxE tests, to permit GxE VC test scalable to biobank data. SEAGLE employs modern matrix computations to achieve the same â exactâ results as the original GxE VC tests, and does not impose additional assumptions nor relies on approximations. SEAGLE can easily accommodate sample sizes in the order of 10^5, is implementable on standard laptops, and does not require specialized equipment. The accompanying manuscript for this package can be found at Chi, Ipsen, Hsiao, Lin, Wang, Lee, Lu, and Tzeng. (2021+) <arXiv:2105.03228>.

Alternative to using withCallingHandlers() in the simple case of catch and rethrow. The `%!%` operator evaluates the expression on its left hand side, and if an error occurs, the right hand side is used to construct a new error that embeds the original error.

This package implements an extension of the Generalized Berk-Jones (GBJ) statistic for survival data, sGBJ. It computes the sGBJ statistic and its p-value for testing the association between a gene set and a time-to-event outcome with possible adjustment on additional covariates. Detailed method is available at Villain L, Ferte T, Thiebaut R and Hejblum BP (2021) <doi:10.1101/2021.09.07.459329>.

Stores and eases the manipulation of spectra and associated data, with dedicated classes for spatial and soil-related data.