Marina Gomtsyan

Postdoctoral researcher in statistics, working on the analysis of spatial inequalities in intergenerational social mobility in France, with Aurélie Fischer and Cyril Jayet.

I completed my PhD in MIA Paris-Saclay (AgroParisTech, University of Paris-Saclay) under the supervision of Céline Lévy-Leduc, Sarah Ouadah and Laure Sansonnet. I worked on variable selection methods in sparse GLARMA models.

• Variable selection, sparsity, regularised methods
• Time series, count data, dependencies
• Contingency tables, categorical data
• Applications in genomics and social sciences

• C. Jayet, M. Gomtsyan, A. Fischer, F. Gargiulo, M. Lenormand (2025). The geography of social class mobility in France : a multidimensional approach. Submitted
• M. Gomtsyan, C. Lévy-Leduc, S. Ouadah, L. Sansonnet, C. Bailly, L. Rajjou (2025). Variable selection in sparse multivariate GLARMA models: application to germination control by environment. To appear in Statistical Methods & Applications
• M. Gomtsyan, C. Lévy-Leduc, S. Ouadah, L. Sansonnet (2024). Sign-consistent estimation in a sparse Poisson model. Statistics & Probability Letters 209, 110107
• M. Gomtsyan (2024). Variable selection in a specific regression time series of counts. arXiv:2307.00929
• M. Gomtsyan, C. Lévy-Leduc, S. Ouadah, L. Sansonnet, T. Blein (2022). Variable selection in sparse GLARMA models. Statistics 56(4), 755–784
• M. Gomtsyan, N. Mokrov, M. Panov, Y. Yanovich (2019). Geometry-Aware Maximum Likelihood Estimation of Intrinsic Dimension. Proceedings of The Eleventh Asian Conference on Machine Learning, PMLR 101:1126-1141

• M. Gomtsyan (2023). NBtsVarSel: Variable Selection in a Specific Regression Time Series of Counts
• M. Gomtsyan (2022). MultiGlarmaVarSel: Variable selection in sparse multivariate GLARMA models
• M. Gomtsyan, C. Lévy-Leduc, S. Ouadah, L. Sansonnet (2021). GlarmaVarSel: Variable selection in sparse GLARMA models

For more information: CV.