Pircalabelu, E. & Artemiou, A.
The LassoPSVM approach for sufficient dimension reduction using principal projections (Under review)
-
I am a Lecturer (Chargé de cours) at UC Louvain working at the Institute of Statistics, Biostatistics and Actuarial Sciences within LIDAM at the Faculty of Science.
Prior to moving at UC Louvain, I held a Visiting professor position at Ghent University affiliated with the Department of Applied Mathematics, Computer Science and Statistics within the Faculty of Sciences and a Postdoctoral position at KU Leuven affiliated with the ORSTAT department within the Faculty of Economics and Business.
My research focuses on: Models for high-dimensional data, Probabilistic graphical models, Social network models, Copula models and Information criteria.
I am currently teaching
Technical reports
Pircalabelu, E. & Artemiou, A.
Graph informed sufficient dimension reduction. (Under review)
Pircalabelu, E. & Claeskens, G.
Linear manifold modeling and graph estimation based on multivariate functional data. (In preparation)
Pircalabelu, E. & Mohammadi, A.
Functional connectivity estimation using informative prior knowledge. (In preparation)
Publications
Pircalabelu, E. & Claeskens, G. (2020).
"Community-Based Group Graphical Lasso", Journal of Machine Learning Research, Accepted.
Pircalabelu, E., Claeskens, G. & Waldorp, L.J. (2019)
"Zoom-in/out joint graphical lasso for different coarseness scales." Journal of the Royal Statistical Society (C), 69(1), 47-67
The final publication is available at https://rss.onlinelibrary.wiley.com/doi/10.1111/rssc.12378.An online supplement is available at https://rss.onlinelibrary.wiley.com/action/downloadSupplement?doi=10.1111%2Frssc.12378&file=rssc12378-sup-0001-Supinfo.pdf
Pircalabelu, E., Claeskens, G. & Gijbels, I. (2017).
"Copula directed acyclic graphs", Statistics and Computing, 27(1), 55-78
The final publication is available at http://link.springer.com/article/10.1007/s11222-015-9599-9Pircalabelu, E., Claeskens, G. & Waldorp, L.J. (2016) .
"Mixed scale joint graphical lasso", Biostatistics, 17(4), 793-806.
The final publication is available at http://biostatistics.oxfordjournals.org/content/early/2016/06/17/biostatistics.kxw025.full.pdf+htmlPircalabelu, E. & Claeskens, G. (2016).
"Focused model selection for social networks", Social Networks, 46, 76-86
The final publication is available at http://www.sciencedirect.com/science/article/pii/S0378873315301763Pircalabelu, E., Claeskens, G., Jahfari, S. & Waldorp, L.J. (2015).
"A focused information criterion for graphical models in fMRI connectivity with high-dimensional data", The Annals of Applied Statistics, 9(4), 2179-2214
The final publication is available at http://dx.doi.org/10.1214/15-AOAS882Pircalabelu, E., Claeskens, G. & Waldorp, L.J. (2015).
"A focused information criterion for graphical models", Statistics and Computing, 25(6), 1071-1092
The final publication is available at http://dx.doi.org/10.1007/s11222-014-9504-yProceedings
Pircalabelu, E., Claeskens, G., & Waldorp, L.J. (2017).
"Top-down joint graphical lasso."
Proceedings of the 32nd International Workshop on Statistical Modelling, July 2-7th 2017,
Groningen. University of Groningen. Pages 47-50.
The final publication is available at
https://iwsm2017.webhosting.rug.nl/IWSM_2017_V2.pdf
Pircalabelu, E., Claeskens, G., Jahfari, S. & Waldorp, L.J. (2014).
"Nodewise graphical modeling using the Focused Information Criterion for 'p larger than n' settings."
Proceedings of the 29th International Workshop on Statistical Modelling, July 14-18th 2014, Göttingen. Georg-August-Universität Göttingen. Pages 273-278.
The final publication is available at
http://www.statmod.org/files/proceedings/iwsm2014_proceedings_vol1.pdf
Book chapters
Claeskens, G., Pircalabelu, E. & Waldorp, L.J.
"Constructing Graphical Models via the Focused Information Criterion."
In: Lecture Notes in Statistics: Modeling and Stochastic Learning for Forecasting in High Dimension (2015).
Editors: A. Antoniadis, X. Brossat, J.-M. Poggi. Pages 55-78.
The final publication is available at link.springer.com via
http://www.springer.com/us/book/9783319187310
Involvement
As part of the open-science paradigme that UCLouvain strongly believes in, I am currently together with other colleagues at UCLouvain, developing the RShiny@UCLouvain platform that will host didactical apps created in Shiny for classroom usage.
COVID - 19
https://eugenpircalabelu.shinyapps.io/covid19-forecasting .
More on the subject here: https://uclouvain.be/fr/instituts-recherche/lidam/actualites/how-can-we-predict-the-evolution-of-covid-19-in-belgium.html .
22/04/2020: Our analysis shows that the number of hospitalized patients and patiencts in ICU is on slightly decaying trends. For the number of deaths, the model still foresees increments for the coming period.
15/04/2020: Our analysis shows that the number of hospitalized patients and patiencts in ICU is stabilizing. Regarding the number of deaths, the model still foresees increments for the coming period.
10/04/2020: Our analysis shows that the number of hospitalized patients is reducing, while the number of patients in ICU is stablizing. Due to the latests developments regarding the reporting in nursing homes and recent additions of extra cases that were not reported previously, a model suggesting a steep growth rate is still plausible.
06/04/2020: Our analysis shows that the number of people hospitalized continues to stabilize. The model predicted for the first time a value lower than the one predicted the day before. The number of ICU patients seems to stabilize as well. The number of deaths has probably exited an exponential growth regime and entered a linear increment phase, which implies that for the following days the number of deaths will most probably still increase. Relative to a SIR model the observed incidence is lower, suggesting the effect of the lockdown measures is now visible in hindering the infections.
01/04/2020: Our analysis shows that the time effect seems to dampen for the number of people hospitalized, resulting in a more stable number of new people being hospitalized in the coming period. The number of ICU patients seems to have exited an exponential regime and entered a linear increment faze in the coming period. As the ICU capacity is limited, one still needs a close monitoring of this indicator in the coming days. The number of deaths still follows an exponential rate at this point in time, which implies that for the following days the number of deaths will most probably still raise.
Contact
Room d.125, 1st floor; Voie du Roman Pays 20, 1348 Louvain-la-Neuve, Belgium