A5084745161- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Sensory Analysis and Statistical Methods
- Advanced Causal Inference Techniques
- Bayesian Methods and Mixture Models
- Advanced Statistical Methods and Models
- Sparse and Compressive Sensing Techniques
- Blind Source Separation Techniques
- Statistical Methods and Applications
- Face and Expression Recognition
- Neural Networks and Applications
- Bayesian Modeling and Causal Inference
- Data Analysis with R
- Spectroscopy and Chemometric Analyses
- Statistical and numerical algorithms
- Statistical Methods in Clinical Trials
- Trauma and Emergency Care Studies
- Gaussian Processes and Bayesian Inference
- Advanced Clustering Algorithms Research
- Machine Learning in Healthcare
- Data Mining Algorithms and Applications
- Data Management and Algorithms
- Direction-of-Arrival Estimation Techniques
- Hydrology and Drought Analysis
- Trauma, Hemostasis, Coagulopathy, Resuscitation
Centre de Mathématiques Appliquées
2016-2024
École Polytechnique
2016-2024
Institut national de recherche en informatique et en automatique
2006-2024
Université de Montpellier
2019-2024
Université Paris-Sud
2016-2024
Neuropsychiatrie : Recherche Epidemiologique et Clinique
2020-2024
Institut Universitaire de Recherche Clinique
2021-2024
Capgemini (Netherlands)
2024
Invent (Germany)
2024
Centre Hospitalier Universitaire de Grenoble
2024
In this article, we present <b>FactoMineR</b> an R package dedicated to multivariate data analysis. The main features of is the possibility take into account different types variables (quantitative or categorical), structure on (a partition variables, a hierarchy individuals) and finally supplementary information (supplementary individuals variables). Moreover, dimensions issued from exploratory analyses can be automatically described by quantitative and/or categorical variables. Numerous...
We present the R package missMDA which performs principal component methods on incomplete data sets, aiming to obtain scores, loadings and graphical representations despite missing values. Package include analysis for continuous variables, multiple correspondence categorical factorial mixed both factor multi-table data. Furthermore, can be used perform single imputation complete involving continuous, variables. A method is also available. In framework, variability across different...
Simple correlation coefficients between two variables have been generalized to measure association matrices in many ways. Coefficients such as the RV coefficient, distance covariance (dCov) coefficient and kernel based are being used by different research communities. Scientists use these test whether random vectors linked. Once it has ascertained that there is through testing, then a next step, often ignored, explore uncover association's underlying patterns. This article provides survey of...
We propose a multiple imputation method based on principal component analysis (PCA) to deal with incomplete continuous data. To reflect the uncertainty of parameters from one next, we use Bayesian treatment PCA model. Using simulation study and real data sets, is compared two classical approaches: joint modelling fully conditional modelling. Contrary others, proposed can be easily used sets where number individuals less than variables when are highly correlated. In addition, it provides...
In many application settings, the data have missing entries which make analysis challenging. An abundant literature addresses values in an inferential framework: estimating parameters and their variance from incomplete tables. Here, we consider supervised-learning settings: predicting a target when appear both training testing data. We show consistency of two approaches prediction. A striking result is that widely-used method imputing with constant, such as mean prior to learning consistent...
In principal component analysis (PCA), the first few components possibly reveal interesting systematic patterns in data, whereas last may reflect random noise. The researcher wonder how many are statistically significant. Many methods have been proposed for determining to retain model, but most of these assume non-standardized data. agricultural, biological and environmental applications, however, standardization is often required. This article proposes parametric bootstrap hypothesis...
Missing values are unavoidable when working with data. Their occurrence is exacerbated as more data from different sources become available. However, most statistical models and visualization methods require complete data, improper handling of missing results in information loss or biased analyses. Since the seminal work Rubin (1976), a burgeoning literature on has arisen, heterogeneous aims motivations. This led to development various methods, formalizations, tools. For practitioners, it...
Abstract Background As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values. These large are well suited train machine learning models, e.g., for forecasting or extract biomarkers in biomedical settings. Such predictive approaches can use discriminative—rather than generative—modeling thus open the door new missing-values strategies. Yet existing empirical evaluations of strategies handle values have focused on inferential...
Abstract Background Impact of in-ICU transfusion on long-term outcomes remains unknown. The purpose this study was to assess in critical-care survivors the association between red blood cells and 1-year mortality. Methods FROG-ICU, a multicenter European enrolling all-comers critical care patients analyzed ( n = 1551). Association administered intensive unit mortality using an augmented inverse probability treatment weighting-augmented censoring weighting method control confounders. Results...