Christine De Mol

Overview

Christine De Mol is an applied mathematician (Université Libre de Bruxelles / ECARES), working on inverse problems, sparse and $\ell_1$ regularization, and their applications in signal processing and econometric forecasting.

Key Contributions / Features

• Iterative soft-thresholding for linear inverse problems with sparsity constraints (Daubechies–Defrise–De Mol 2004).
• Sparse and stable Markowitz portfolios (Brodie et al. 2009): $\ell_1$-penalized portfolio selection.
• Bayesian shrinkage vs principal components (De Mol-Giannone-Reichlin 2008): forecasting with many predictors via regularization — ridge ≈ principal components, lasso for sparsity (Dynamic Factor Model).

Related

• Sparse Portfolio Selection
• Dynamic Factor Model
• Lasso
• De Mol-Giannone-Reichlin (2008) Forecasting Using a Large Number of Predictors
• Brodie-Daubechies-De Mol-Giannone-Loris (2009) Sparse and Stable Markowitz Portfolios
• Ingrid Daubechies
• Domenico Giannone