Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
Predicting colitis using penalised regression
Published:
Wrote an R tutorial for Imperial College students that demonstrates how penalised regression can be used to predict cases of colitis, using gene expression data.
Which Premier League team has the most impactful home fans? (Significance magazine updated version)
Published:
I previously conducted a statistical analysis into whether home fans provide an advantage in the English Premier League teams (see Towards Data Science). I have since updated and expanded the analysis and was selected as one of the finalists for the 2024 Significance magazine writing competition. The story can be found on the Significance website.
Which Premier League team has the most impactful home fans?
Published:
Conducted a statistical analysis into whether home fans provide an advantage and which English Premier League teams benefit from the most ‘impactful’ fans. It was posted on Towards Data Science.
portfolio
sgs R package
An R package to fit sparse-group SLOPE (SGS) models.
dfr R package
An R package to fit sparse-group lasso (SGL) models using the dual feature reduction (DFR) approach.
publications
Sparse-group SLOPE: adaptive bi-level selection with FDR-control
Released in 2023
This paper presents a new high-dimensional approach for simultaneous variable and group selection, called Sparse-group SLOPE (SGS).
Recommended citation: Fabio Feser and Marina Evangelou (2023). "Sparse-group SLOPE: adaptive bi-level selection with FDR-control". arXiv preprint arXiv:2305.09467.
Download Paper
Strong screening rules for group-based SLOPE models
Released in 2024
This paper presents strong screening rules for group SLOPE and sparse-group SLOPE. Accepted at AISTATS 25.
Recommended citation: Fabio Feser and Marina Evangelou (2025). "Strong screening rules for group-based SLOPE models". AISTATS 25, PMLR 258:352-360, 2025.
Download Paper
Dual feature reduction for the sparse-group lasso and its adaptive variant
Released in 2024
This paper presents a new feature reduction approach for the sparse-group lasso and adaptive sparse-group lasso, called Dual Feature Reduction (DFR). Accepted at ICML 25.
Recommended citation: Fabio Feser and Marina Evangelou (2024). "Dual feature reduction for the sparse-group lasso and its adaptive variant". arXiv preprint arXiv:2405.17094.
Download Paper
talks
Controlling the FDR in high-dimensional settings
Published:
Gave a talk on approaches for controlling the false-discovery rate (FDR) under high-dimensional settings. The talk was given to the Mary Lister scholars at Imperial College.
Bi-level variable and group selection in genetics
Published:
Discussed various approaches towards bi-level selection in a genetics framework, including sparse-group methods such as the sparse-group lasso and sparse-group SLOPE.
Sparse-group SLOPE: adaptive bi-level selection with FDR-control
Published:
Presented sparse-group SLOPE (SGS) at the CMStatistics 2023 conference.
Disease prediction using pathway information
Published:
Presented a lecture on using penalised regression to perform disease prediction, as part of the Statistical Genetics module on the MSc Statistics at Imperial College.
Sparse-group SLOPE: Adaptive bi-level selection with FDR-control
Published:
Invited talk to present ‘Sparse-group SLOPE: Adaptive bi-level selection with FDR-control’ during the ‘Reliable prediction models for challenging data’ session at COMPSTAT 2024.
High-dimensional sparse-group penalised regression models
Published:
Presented a talk on adaptive sparse-group models at the UCPH Statistics Seminar at the University of Copenhagen.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.