FLR algorithm for classification.

Lognormal models have broad applications in various research areas such as economics, actuarial science, biology, environmental science and psychology. The estimation problem in lognormal models has been extensively studied. This R package fuel implements thirty-nine existing and newly proposed estimators. See Zhang, F., and Gou, J. (2020), A unified framework for estimation in lognormal models, Technical report.

Given vectors of family sizes and number of affecteds per family, calculates the risk of disease recurrence in an unaffected person, conditional on a family having at least k affected members. Methods also model heterogeneity of disease risk across families by fitting a mixture model, allowing for high and low risk families.

We propose an objective Bayesian algorithm for searching the space of Gaussian directed acyclic graph (DAG) models. The algorithm uses moment fractional Bayes factors (MFBF) and is suitable for learning sparse graphs. The algorithm is implemented using Armadillo, an open-source C++ linear algebra library.

This package performs functional regression, and some related approaches, for intensive longitudinal data (see the book by Walls & Schafer, 2006, Models for Intensive Longitudinal Data, Oxford) when such data is not necessarily observed on an equally spaced grid of times. The approach generally follows the ideas of Goldsmith, Bobb, Crainiceanu, Caffo, and Reich (2011)<DOI:10.1198/jcgs.2010.10007> and the approach taken in their sample code, but with some modifications to make it more feasible to use with long rather than wide, non-rectangular longitudinal datasets with unequal and potentially random measurement times. It also allows easy plotting of the correlation between the smoothed covariate and the outcome as a function of time, which can add additional insights on how to interpret a functional regression. Additionally, it also provides several permutation tests for the significance of the functional predictor. The heuristic interpretation of ``time is used to describe the index of the functional predictor, but the same methods can equally be used for another unidimensional continuous index, such as space along a north-south axis. Note that most of the functionality of this package has been superseded by added features after 2016 in the pfr function by Jonathan Gellar, Mathew W. McLean, Jeff Goldsmith, and Fabian Scheipl, in the refund package built by Jeff Goldsmith and co-authors and maintained by Julia Wrobel. The development of the funreg package in 2015 and 2016 was part of a research project supported by Award R03 CA171809-01 from the National Cancer Institute and Award P50 DA010075 from the National Institute on Drug Abuse. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse, the National Cancer Institute, or the National Institutes of Health.

Calculate useful quantities for a user-defined differential equation model of infectious disease transmission among individuals in a healthcare facility. Input rates of transition between states of individuals with and without the disease-causing organism, distributions of states at facility admission, relative infectivity of transmissible states, and the facility length of stay distribution. Calculate the model equilibrium and the basic facility reproduction number, as described in Toth et al. (2025) <doi:10.1371/journal.pcbi.1013577>.

Fair machine learning regression models which take sensitive attributes into account in model estimation. Currently implementing Komiyama et al. (2018) <http://proceedings.mlr.press/v80/komiyama18a/komiyama18a.pdf>, Zafar et al. (2019) <https://www.jmlr.org/papers/volume20/18-262/18-262.pdf> and my own approach from Scutari, Panero and Proissl (2022) <doi:10.1007/s11222-022-10143-w> that uses ridge regression to enforce fairness.

Comparisons of floating point numbers are problematic due to errors associated with the binary representation of decimal numbers. Despite being aware of these problems, people still use numerical methods that fail to account for these and other rounding errors (this pitfall is the first to be highlighted in Circle 1 of Burns (2012) The R Inferno <https://www.burns-stat.com/pages/Tutor/R_inferno.pdf>). This package provides new relational operators useful for performing floating point number comparisons with a set tolerance.

This package contains a set of functions that can be used to apply formats to data frames or vectors. The package aims to provide functionality similar to that of SAS® formats. Formats are assigned to the format attribute on data frame columns. Then when the fdata() function is called, a new data frame is created with the column data formatted as specified. The package also contains a value() function to create a user-defined format, similar to a SAS® user-defined format.

Generate regression results tables and plots in final format for publication. Explore models and export directly to PDF and Word using RMarkdown'.

Implementation of the Interval Testing Procedure for functional data in different frameworks (i.e., one or two-population frameworks, functional linear models) by means of different basis expansions (i.e., B-spline, Fourier, and phase-amplitude Fourier). The current version of the package requires functional data evaluated on a uniform grid; it automatically projects each function on a chosen functional basis; it performs the entire family of multivariate tests; and, finally, it provides the matrix of the p-values of the previous tests and the vector of the corrected p-values. The functional basis, the coupled or uncoupled scenario, and the kind of test can be chosen by the user. The package provides also a plotting function creating a graphical output of the procedure: the p-value heat-map, the plot of the corrected p-values, and the plot of the functional data.

Implementation of dynamic principal component analysis (DPCA), simulation of VAR and VMA processes and frequency domain tools. These frequency domain methods for dimensionality reduction of multivariate time series were introduced by David Brillinger in his book Time Series (1974). We follow implementation guidelines as described in Hormann, Kidzinski and Hallin (2016), Dynamic Functional Principal Component <doi:10.1111/rssb.12076>.

Fits a functional mediation model with a scalar distal outcome. The method is described in detail by Coffman, Dziak, Litson, Chakraborti, Piper & Li (2021) <arXiv:2112.03960>. The model is similar to that of Lindquist (2012) <doi:10.1080/01621459.2012.695640> although allowing a binary outcome as an alternative to a numerical outcome. The current version is a minor bug fix in the vignette. The development of this package was part of a research project supported by National Institutes of Health grants P50 DA039838 from the National Institute of Drug Abuse and 1R01 CA229542-01 from the National Cancer Institute and the NIH Office of Behavioral and Social Science Research. Content is solely the responsibility of the authors and does not necessarily represent the official views of the funding institutions mentioned above. This software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

Fast and flexible Kalman filtering and smoothing implementation utilizing sequential processing, designed for efficient parameter estimation through maximum likelihood estimation. Sequential processing is a univariate treatment of a multivariate series of observations and can benefit from computational efficiency over traditional Kalman filtering when independence is assumed in the variance of the disturbances of the measurement equation. Sequential processing is described in the textbook of Durbin and Koopman (2001, ISBN:978-0-19-964117-8). FKF.SP was built upon the existing FKF package and is, in general, a faster Kalman filter/smoother.

Implement and fit a variety of short-memory (SM) and long-memory (LM) models from a very broad family of exponential generalized autoregressive conditional heteroskedasticity (EGARCH) models, such as a MEGARCH (modified EGARCH), FIEGARCH (fractionally integrated EGARCH), FIMLog-GARCH (fractionally integrated modulus Log-GARCH), and more. The FIMLog-GARCH as part of the EGARCH family is discussed in Feng et al. (2023) <https://econpapers.repec.org/paper/pdnciepap/156.htm>. For convenience and the purpose of comparison, a variety of other popular SM and LM GARCH-type models, like an APARCH model, a fractionally integrated APARCH (FIAPARCH) model, standard GARCH and fractionally integrated GARCH (FIGARCH) models, GJR-GARCH and FIGJR-GARCH models, TGARCH and FITGARCH models, are implemented as well as dual models with simultaneous modelling of the mean, including dual long-memory models with a fractionally integrated autoregressive moving average (FARIMA) model in the mean and a long-memory model in the variance, and semiparametric volatility model extensions. Parametric models and parametric model parts are fitted through quasi-maximum-likelihood estimation. Furthermore, common forecasting and backtesting functions for value-at-risk (VaR) and expected shortfall (ES) based on the package's models are provided.

This package provides a collection of user-friendly functions for assessing and visualizing fragility of individual studies (Walsh et al., 2014 <doi:10.1016/j.jclinepi.2013.10.019>; Lin, 2021 <doi:10.1111/jep.13428>), conventional pairwise meta-analyses (Atal et al., 2019 <doi:10.1016/j.jclinepi.2019.03.012>), and network meta-analyses of multiple treatments with binary outcomes (Xing et al., 2020 <doi:10.1016/j.jclinepi.2020.07.003>). The included functions are designed to: 1) calculate the fragility index (i.e., the minimal event status modifications that can alter the significance or non-significance of the original result) and fragility quotient (i.e., fragility index divided by sample size) at a specific significance level; 2) give the cases of event status modifications for altering the result's significance or non-significance and visualize these cases; 3) visualize the trend of statistical significance as event status is modified; 4) efficiently derive fragility indexes and fragility quotients at multiple significance levels, and visualize the relationship between these fragility measures against the significance levels; and 5) calculate fragility indexes and fragility quotients of multiple datasets (e.g., a collection of clinical trials or meta-analyses) and produce plots of their overall distributions. The outputs from these functions may inform the robustness of clinical results in terms of statistical significance and aid the interpretation of fragility measures. The usage of this package is illustrated in Lin et al. (2023 <doi:10.1016/j.ajog.2022.08.053>) and detailed in Lin and Chu (2022 <doi:10.1371/journal.pone.0268754>).

Stores large arrays in files to avoid occupying large memories. Implemented with super fast gigabyte-level multi-threaded reading/writing via OpenMP'. Supports multiple non-character data types (double, float, complex, integer, logical, and raw).

Fit growth models to otoliths and/or tagging data, using the RTMB package and maximum likelihood. The otoliths (or similar measurements of age) provide direct observed coordinates of age and length. The tagging data provide information about the observed length at release and length at recapture at a later time, where the age at release is unknown and estimated as a vector of parameters. The growth models provided by this package can be fitted to otoliths only, tagging data only, or a combination of the two. Growth variability can be modelled as constant or increasing with length.

Edit vectors to fill missing values, based on the vector itself.

Scrapes data from Fitbit <http://www.fitbit.com>. This does not use the official API, but instead uses the API that the web dashboard uses to generate the graphs displayed on the dashboard after login at <http://www.fitbit.com>.

This package provides a utility to scrape and load play-by-play data and statistics from the Premier Hockey Federation (PHF) <https://www.premierhockeyfederation.com/>, formerly known as the National Women's Hockey League (NWHL). Additionally, allows access to the National Hockey League's stats API <https://www.nhl.com/>.

Interface to Palantir Foundry', including reading and writing structured or unstructured datasets, and more <https://www.palantir.com/platforms/foundry/>.

This package provides functions that support stable prediction and classification with radiomics data through factor-analytic modeling. For details, see Peeters et al. (2019) <doi:10.48550/arXiv.1903.11696>.

Genotyping assays for bi-allelic markers (e.g. SNPs) produce signal intensities for the two alleles. fitPoly assigns genotypes (allele dosages) to a collection of polyploid samples based on these signal intensities. fitPoly replaces the older package fitTetra that was limited (a.o.) to only tetraploid populations whereas fitPoly accepts any ploidy level. Reference: Voorrips RE, Gort G, Vosman B (2011) <doi:10.1186/1471-2105-12-172>. New functions added on conversion of data from SNP array software formats, drawing of XY-scatterplots with or without genotype colors, checking against expected F1 segregation patterns, comparing results from two different assays (probes) for the same SNP, recovery from a saveMarkerModels() crash.