Software

  • R package roahd - (RObust Analysis of High dimensional Data)

Link: roahd manual on CRAN

Authors: Nicholas Tarabelloni (mantainer), Ana Arribas-Gil, Francesca Ieva, Anna M. Paganoni, Juan Romo

Description: roahd is an R package meant to gather recently proposed statistical methods to deal with the robust analysis of functional data. The package contains an implementation of quantitative methods, based on functional depths and indices, on top of which are built some graphical methods (functional boxplot and outliergram) useful to carry out an explorative analysis of a functional dataset and to robustify it by discarding shape and magnitude outliers. Both univariate and multivariate functional data are supported in the package, whose functions have been implemented with a particular emphasis on computational efficiency, in order to allow the processing of high-dimensional dataset. Tools for evaluating dependency (and patterns of dependency) among families of (multivariate) curves through Spearman index and Matrices and for making inference on that in a bootstrap based context are also provided.

For further details, see Ieva et al. (2018) [url], Ieva and Paganoni (2017) [url], Tarabelloni et al. (2016) [url], Ieva and Paganoni (2013).

Note: built with R 3.4.4


  • R package discfrail - Cox Models for Time-to-Event Data with Nonparametric Discrete Group-Specific Frailties

Link: discfrail on github

Authors: Francesca Gasperoni, Christopher Jackson

Description: The package contains functions for fitting Cox proportional hazards models for grouped time-to-event data, where the shared group-specific frailties have a discrete nonparametric distribution. Functions for simulating from these models, with a nonparametric or a parametric baseline hazard function are also present.

The methods proposed in the package is described by Gasperoni, F. et al. (2018) [url].

Note: built with R 3.4.4


  • R package BLSM - Bayesian Latent Space Models

Link: BLSM manual on CRAN

Author: Alberto Donizetti (mantainer)

Supervisor: Francesca Ieva

Description: BLSM is an R package meant to allow the computation of an improved version of Bayesian Latent Space Model for Complex Networks, with respect to the one proposed in Hoff (2002). The goal of Latent Space Models is to map the observed network in the latent space by meeting precise probabilistic requirements: once the latent positions have been estimated, different kinds of clustering can be performed in the latent space to group similar nodes as an interesting alternative to algorithms designed for community detection.

For further details, see here.

Note: built with R 3.4.4


  • R package msmtools - Building Augmented Data to Run Multi-State Models with 'msm'

Link: msmtools manual on CRAN

Author: Francesco Grossetti (mantainer)

Supervisor: Francesca Ieva, Anna M. Paganoni

Description: msmtools is an R package meant to provide a fast and general method for restructuring classical longitudinal data into augmented ones. The reason for this is to facilitate the modeling of longitudinal data under a multi-state framework using the 'msm' package.

For further details, see Grossetti et al. (2017) [url], Ieva et al. (2017) [url], Gasperoni et al. (2017) [url].

Note: built with R 3.2.5


  • R package gmfd - Inference and Clustering of Functional Data

Link: gmfd manual on CRAN

Authors: Andrea Martino (mantainer), Andrea Ghiglietti, Francesca Ieva, Anna M. Paganoni

Description: gmfd is an R package meant to provide some methods for the inference and clustering of univariate and multivariate functional data, using a generalization of Mahalanobis distance, along with some functions

useful for the analysis of functional data.

For further details, see Martino A., Ghiglietti, A., Ieva, F. and Paganoni A. M. (2017) [url].

Note: built with R 3.2.5