# Valentin Todorov

#### 13 packages on CRAN

Provides interface to the 'MATLAB' toolbox 'Flexible Statistical Data Analysis (FSDA)' which is a comprehensive and computationally efficient software package for robust statistics in regression, multivariate and categorical data analysis. The current R version implements tools for regression: (forward search, S- and MM-estimation, least trimmed squares (LTS) and least median of squares (LMS)), for multivariate analysis (forward search, S- and MM-estimation), for cluster analysis and cluster-wise regression. The distinctive feature of our package is the possibility of monitoring the statistics of interest as function of breakdown point, efficiency or subset size, depending on the estimator. This is accompanied by a rich set of graphical features, such as dynamic brushing, linking, particularly useful for exploratory data analysis.

Provides functions for robust PCA by projection pursuit. The methods are described in Croux et al. (2006) <doi:10.2139/ssrn.968376>, Croux et al. (2013) <doi:10.1080/00401706.2012.727746>, Todorov and Filzmoser (2013) <doi:10.1007/978-3-642-33042-1_31>.

Methods for robust statistics, a state of the art in the early 2000s, notably for robust regression and robust multivariate analysis.

Robust Location and Scatter Estimation and Robust Multivariate Analysis with High Breakdown Point: principal component analysis (Filzmoser and Todorov (2013), <doi:10.1016/j.ins.2012.10.017>), linear and quadratic discriminant analysis (Todorov and Pires (2007)), multivariate tests (Todorov and Filzmoser (2010) <doi:10.1016/j.csda.2009.08.015>), outlier detection (Todorov et al. (2010) <doi:10.1007/s11634-010-0075-2>). See also Todorov and Filzmoser (2009) <ISBN-13:978-3838108148>, Todorov and Filzmoser (2010) <doi:10.18637/jss.v032.i03> and Boudt et al. (2019) <doi:10.1007/s11222-019-09869-x>.

Provides methods for multiway data analysis by means of Parafac and Tucker 3 models. Robust versions (Engelen and Hubert (2011) <doi:10.1016/j.aca.2011.04.043>) and versions for compositional data are also provided (Gallo (2015) <doi:10.1080/03610926.2013.798664>, Di Palma et al. (2018) <doi:10.1080/02664763.2017.1381669>.

Robust multivariate methods for high dimensional data including outlier detection, PCA, PLS and classification.

Robust Location and Scatter Estimation and Robust Multivariate Analysis with High Breakdown Point for Incomplete Data (missing values) (Todorov et al. (2010) <doi:10.1007/s11634-010-0075-2>).

Provides functions for fitting a sparse partial least squares (SPLS) regression and classification (Chun and Keles (2010) <doi:10.1111/j.1467-9868.2009.00723.x>).

Provides functions for robust trimmed clustering. The methods are described in Garcia-Escudero (2008) <doi:10.1214/07-AOS515>, Fritz et al. (2012) <doi:10.18637/jss.v047.i12> and others.

Trimmed k-means clustering. The method is described in Cuesta-Albertos et al. (1997) <doi:10.1214/aos/1031833664>.

"Essential" Robust Statistics. Tools allowing to analyze data with robust methods. This includes regression methodology including model selections and multivariate statistics where we strive to cover the book "Robust Statistics, Theory and Methods" by 'Maronna, Martin and Yohai'; Wiley 2006.

Implementation of robust sparse PCA using the ROSPCA algorithm of Hubert et al. (2016) <DOI:10.1080/00401706.2015.1093962>.