PhD in Statistics, 2020
University of Bologna
MSc in Statistical Science, 2016
University of Bologna
BSc in Statistics (Double Degree), 2014
University of Bologna, University of Glasgow
R, Python, SPSS,
SAS, SQL, Git
R Markdown, Power BI,
LaTeX, beamer
Innovation, Fast learning,
Adaptability, Commitment,
Public Speaking, Technical Writing
Classification of the operational states of a wastewater treatment plant through clustering (hierarchical, partitioning, model-based, and modal clustering), hidden Markov models, and dimension reduction methods.
Exploration of the best hip hop songs, analysis of their audio features obtained from the Spotify API, exploratory factor analysis and dimension reduction.
Apply mixture simultaneous factor analysis and hypothesis testing to classify countries on the basis of the cultural appropriateness of particular emotions.
Non-negative matrix factorization for optimal compression of the painting Madonna del Cardellino through a low-rank approximation of the pixel intensities matrix.
Data analysis of the characteristics of the artwork pieces displayed at MoMA (New York).
Codes and tools to estimate penalized factor models through approximations of convex (e.g., lasso, adaptive lasso) and non-convex (e.g., scad, mcp) penalties.
Personal notebooks and analyses available at Kaggle, the Data Science and Machine Learning Community.
Assessment of students’ psychological abilities through a properly estimated penalized multiple-group factor analysis model.
Recover the latent topics in the complete works by Shakespeare and classify the document corpus accordingly.
Exploration of the most popular characters, locations, and guest stars, analysis of TV ratings, votes and episodes views, text analysis on the dialogues, sentiment analysis, topic modelling.
Social Statistics Seminar at the London School of Economics (LSE) on “A penalized likelihood-based approach for single and multiple-group factor analysis models”, London, UK (Nov 2019).
Talk at International Meeting of the Psychometric Society (IMPS) on “Sparse Multigroup Factor Analysis”, Columbia University, New York City, USA (Jul 2018).
ACTNext Tech Talk on “Sparse Multigroup Factor Analysis for Continuous and Categorical Data” at ACTNext, Iowa City, USA (Jul 2018). See post on Twitter.
Geminiani, E., Ceulemans, E., & De Roover, K. (2020). Testing for factor loading differences in mixture simultaneous factor analysis: a Monte Carlo simulation-based perspective. Structural Equation Modeling: A Multidisciplinary Journal, 1-19. DOI: 10.1080/10705511.2020.1807351
Geminiani, E. (2020). A penalized likelihood-based framework for single and multiple-group factor analysis models (Doctoral dissertation, University of Bologna). DOI: 10.6092/unibo/amsdottorato/9355 Available at http://amsdottorato.unibo.it/9355/.