* Estimation with Penalisation in Systems of Ordinary Differential Equations* (

**episode**)

A set of statistical tools for inferring unknown parameters in continuous time processes governed by ordinary differential equations (ODE). Moreover, variance reduction and model selection can be obtained through various implemented penalisation schemes. The package offers two estimation procedures: exact estimation via least squares and a faster approximate estimation via inverse collocation methods. All estimators can handle multiple data sets arising from the same ODE system, but subjected to different interventions.

*(*

**Subsample Winner Algorithm for Variable Selection in Linear Regression with a Large Number of Variables****subsamp**)

This subsample winner algorithm (SWA) for regression with a large-p data (X, Y) selects the important variables (or features) among the p features X in explaining the response Y. The SWA first uses a base procedure, here a linear regression, on each of subsamples randomly drawn from the p variables, and then computes the scores of all features, i.e., the p variables, according to the performance of these features collected in each of the subsample analyses. It then obtains the ‘semifinalist’ of the features based on the resulting scores and determines the ‘finalists’, i.e., the important features, from the ‘semifinalist’. Fan, Sun and Qiao (2017) <http://…/>.

*(*

**A Clustering Method Based on Boosting on Single Attributes****boclust**)

An overlap clustering algorithm for categorical ultra-dimension data.

*(*

**Unimodal and Isotonic L1, L2 and Linf Regression****UniIsoRegression**)

Perform L1 or L2 isotonic and unimodal regression on 1D weighted or unweighted input vector and isotonic regression on 2D weighted or unweighted input vector. It also performs L infinity isotonic and unimodal regression on 1D unweighted input vector. Reference: Quentin F. Stout (2008) <doi:10.1016/j.csda.2008.08.005>. Spouge, J., Wan, H. & Wilbur, W.(2003) <doi:10.1023/A:1023901806339>. Q.F. Stout (2013) <doi:10.1007/s00453-012-9628-4>.

*(*

**Change-Point Estimation for Expensive and High-Dimensional Models****changepointsHD**)

This implements the methods developed in, L. Bybee and Y. Atchade. (2017) <arXiv:1707.04306>. Contains a series of methods for estimating change-points given user specified black-box models. The methods include binary segmentation for multiple change-point estimation. For estimating each individual change-point the package includes simulated annealing, brute force, and, for Gaussian graphical models, an application specific rank-one update implementation. Additionally, code for estimating Gaussian graphical models is included. The goal of this package is to allow for the efficient estimation of change-points in complicated models with high dimensional data.

*(*

**Spatial Dependence Based on Empirical Variograms****variosig**)

Apply Monte Carlo permutation to compute pointwise variogram envelopes, and check spatial dependence using permutation test adjusted for multiple testing.