Applies dynamic structural equation models to time-series data with generic and simplified specification for simultaneous and lagged effects. Methods are described in Thorson et al. (2024) "Dynamic structural equation models synthesize ecosystem dynamics constrained by ecological mechanisms.".

This package provides a statistically and computationally efficient debiasing method for conducting valid inference on the high-dimensional linear regression function with missing outcomes. The reference paper is Zhang, Giessing, and Chen (2023) <doi:10.48550/arXiv.2309.06429>.

This package provides functions to pipe data from R to DataGraph', a graphing and analysis application for mac OS. Create a live connection using either .dtable or .dtbin files that can be read by DataGraph'. Can save a data frame, collection of data frames and sequences of data frames and individual vectors. For more information see <https://community.visualdatatools.com/datagraph/knowledge-base/r-package/>.

This package provides a set of functions to perform distribution-free Bayesian analyses. Included are Bayesian analogues to the frequentist Mann-Whitney U test, the Wilcoxon Signed-Ranks test, Kendall's Tau Rank Correlation Coefficient, Goodman and Kruskal's Gamma, McNemar's Test, the binomial test, the sign test, the median test, as well as distribution-free methods for testing contrasts among condition and for computing Bayes factors for hypotheses. The package also includes procedures to estimate the power of distribution-free Bayesian tests based on data simulations using various probability models for the data. The set of functions provide data analysts with a set of Bayesian procedures that avoids requiring parametric assumptions about measurement error and is robust to problem of extreme outlier scores.

Set of tools aimed at processing meteorological data, converting hourly recorded data to daily, monthly and annual data.

The framework provides functions to generate ODEs of reaction networks, parameter transformations, observation functions, residual functions, etc. The framework follows the paradigm that derivative information should be used for optimization whenever possible. Therefore, all major functions produce and can handle expressions for symbolic derivatives. The methods used in dMod were published in Kaschek et al, 2019, <doi:10.18637/jss.v088.i10>.

Statistical hypothesis testing using the Delta method as proposed by Deng et al. (2018) <doi:10.1145/3219819.3219919>. This method replaces the standard variance estimation formula in the Z-test with an approximate formula derived via the Delta method, which can account for within-user correlation.

Computes the first stage GMM estimate of a dynamic linear model with p lags of the dependent variables.

This package contains a robust set of tools designed for constructing deep neural networks, which are highly adaptable with user-defined loss function and probability models. It includes several practical applications, such as the (deepAFT) model, which utilizes a deep neural network approach to enhance the accelerated failure time (AFT) model for survival data. Another example is the (deepGLM) model that applies deep neural network to the generalized linear model (glm), accommodating data types with continuous, categorical and Poisson distributions.

This package provides a collection of novel tools for generating species distribution and abundance models (SDM) that are dynamic through both space and time. These highly flexible functions incorporate spatial and temporal aspects across key SDM stages; including when cleaning and filtering species occurrence data, generating pseudo-absence records, assessing and correcting sampling biases and autocorrelation, extracting explanatory variables and projecting distribution patterns. Throughout, functions utilise Google Earth Engine and Google Drive to minimise the computing power and storage demands associated with species distribution modelling at high spatio-temporal resolution.

Basic routines used in scientific coding, such as timing routines, vector/array handing functions and I/O support routines.

Fits disaggregation regression models using TMB ('Template Model Builder'). When the response data are aggregated to polygon level but the predictor variables are at a higher resolution, these models can be useful. Regression models with spatial random fields. The package is described in detail in Nandi et al. (2023) <doi:10.18637/jss.v106.i11>.

Fits dose-response models utilizing a Bayesian model averaging approach as outlined in Gould (2019) <doi:10.1002/bimj.201700211> for both continuous and binary responses. Longitudinal dose-response modeling is also supported in a Bayesian model averaging framework as outlined in Payne, Ray, and Thomann (2024) <doi:10.1080/10543406.2023.2292214>. Functions for plotting and calculating various posterior quantities (e.g. posterior mean, quantiles, probability of minimum efficacious dose, etc.) are also implemented. Copyright Eli Lilly and Company (2019).

Model selection algorithms for regression and classification, where the predictors can be continuous or categorical and the number of regressors may exceed the number of observations. The selected model consists of a subset of numerical regressors and partitions of levels of factors. Szymon Nowakowski, Piotr Pokarowski, Wojciech Rejchel and Agnieszka SoÅ tys, 2023. Improving Group Lasso for High-Dimensional Categorical Data. In: Computational Science â ICCS 2023. Lecture Notes in Computer Science, vol 14074, p. 455-470. Springer, Cham. <doi:10.1007/978-3-031-36021-3_47>. Aleksandra Maj-KaÅ ska, Piotr Pokarowski and Agnieszka Prochenka, 2015. Delete or merge regressors for linear model selection. Electronic Journal of Statistics 9(2): 1749-1778. <doi:10.1214/15-EJS1050>. Piotr Pokarowski and Jan Mielniczuk, 2015. Combined l1 and greedy l0 penalized least squares for linear model selection. Journal of Machine Learning Research 16(29): 961-992. <https://www.jmlr.org/papers/volume16/pokarowski15a/pokarowski15a.pdf>. Piotr Pokarowski, Wojciech Rejchel, Agnieszka SoÅ tys, MichaÅ Frej and Jan Mielniczuk, 2022. Improving Lasso for model selection and prediction. Scandinavian Journal of Statistics, 49(2): 831â 863. <doi:10.1111/sjos.12546>.

This package creates interactive genome browser. It joins the data analysis power of R and the visualization libraries of JavaScript in one package. Barrios, D. & Prieto, C. (2017) <doi:10.1089/cmb.2016.0213>.

This package provides a sparse Partial Least Squares implementation which uses soft-threshold estimation of the covariance matrices and therein introduces sparsity. Number of components and regularization coefficients are automatically set.

Calculate posterior modes and credible intervals of parameters of the Dixon-Simon model for subgroup analysis (with binary covariates) in clinical trials. For details of the methodology, please refer to D.O. Dixon and R. Simon (1991), Biometrics, 47: 871-881.

Fit latent variable linear models, estimating score distributions for groups of people, following Cohen and Jiang (1999) <doi:10.2307/2669917>. In this model, a latent distribution is conditional on students item response, item characteristics, and conditioning variables the user includes. This latent trait is then integrated out. This software is intended to fit the same models as the existing software AM <https://am.air.org/>. As of version 2, also allows the user to draw plausible values.

Collection of functions for distributed lag linear and non-linear models.

Estimates the conditional association between an exposure and an outcome given covariates. Three methods are implemented: O-estimation, where a nuisance model for the association between the covariates and the outcome is used; E-estimation where a nuisance model for the association between the covariates and the exposure is used, and doubly robust (DR) estimation where both nuisance models are used. In DR-estimation, the estimates will be consistent when at least one of the nuisance models is correctly specified, not necessarily both. For more information, see Zetterqvist and Sjölander (2015) <doi:10.1515/em-2014-0021>.

Simulates demic diffusion building on models previously developed for the expansion of Neolithic and other food-producing economies during the Holocene (Fort et al. (2012) <doi:10.7183/0002-7316.77.2.203>, Souza et al. (2021) <doi:10.1098/rsif.2021.0499>). Growth and emigration are modelled as density-dependent processes using logistic growth and an asymptotic threshold model. Environmental and terrain layers, which can change over time, affect carrying capacity, growth and mobility. Multiple centres of origin with their respective starting times can be specified.

We consider a multiple testing procedure used in many modern applications which is the q-value method proposed by Storey and Tibshirani (2003), <doi:10.1073/pnas.1530509100>. The q-value method is based on the false discovery rate (FDR), hence versions of the q-value method can be defined depending on which estimator of the proportion of true null hypotheses, p0, is plugged in the FDR estimator. We implement the q-value method based on two classical pi0 estimators, and furthermore, we propose and implement three versions of the q-value method for homogeneous discrete uniform P-values based on pi0 estimators which take into account the discrete distribution of the P-values.

This package performs the drifting Markov models (DMM) which are non-homogeneous Markov models designed for modeling the heterogeneities of sequences in a more flexible way than homogeneous Markov chains or even hidden Markov models. In this context, we developed an R package dedicated to the estimation, simulation and the exact computation of associated reliability of drifting Markov models. The implemented methods are described in Vergne, N. (2008), <doi:10.2202/1544-6115.1326> and Barbu, V.S., Vergne, N. (2019) <doi:10.1007/s11009-018-9682-8> .

Supporting the quantitative analysis of binary welfare based decision making processes using Monte Carlo simulations. Decision support is given on two levels: (i) The actual decision level is to choose between two alternatives under probabilistic uncertainty. This package calculates the optimal decision based on maximizing expected welfare. (ii) The meta decision level is to allocate resources to reduce the uncertainty in the underlying decision problem, i.e to increase the current information to improve the actual decision making process. This problem is dealt with using the Value of Information Analysis. The Expected Value of Information for arbitrary prospective estimates can be calculated as well as Individual Expected Value of Perfect Information. The probabilistic calculations are done via Monte Carlo simulations. This Monte Carlo functionality can be used on its own.