TOMARCHIO Salvatore Daniele 

Nome: Salvatore Daniele 

Cognome: Tomarchio 

Ciclo di dottorato: XXXIII 

I)Personal information: 

Nome: Salvatore Daniele 

Cognome: Tomarchio 

Ciclo di dottorato: XXXIII 

- Personal website: nessuno 

- E-mail: daniele.tomarchio@hotmail.it  

- Current occupation: Assegnista di ricerca presso l'Università degli studi di Catania 

II) Education within the Ph.D: 
-Dissertation title: Advances in finite mixture models:new methodologies and applications 
-Tutor (and eventual co-tutor): Prof. Antonio Punzo 
-Visiting student at : Department of Mathematics and Statistics, McMaster University, Hamilton (ON), Canada, dal 02/01/2019 al 28/06/2019. 
-Participation in courses/summer or winter school: 1) Partecipazione alla scuola estiva “Clustering, Data Analysis and Visualization of Complex Data”, Dipartimento di Economia e Impresa dell’Università degli Studi di Catania, dal 21/05/2018 al 25/05/2018. 

2) Partecipazione alla scuola estiva “Data Science”, Department of Mathematics and Statistics, McMaster University, Hamilton (ON), Canada. Dal 27/05/2019 al 31/05/2019. 
-Area (whether Economic, Business or Statistics): Statistics 
 

III) Other information 
-Research interests: Finite mixture models, Matrix-variate data, Model-based clustering, Hidden Markov Models, Parsimonious Models 
-Grants, Awards: "Honorable Mention" by the Classification Society of North America for the best PhD Thesis, 14/03/2021. 
-Publications: 

A) Articoli su Rivista 

• 2021 Tomarchio S.D., McNicholas P.D. and Punzo A., Matrix Normal Cluster-Weighted Models, Journal of Classification, in press. DOI: https://doi.org/10.1007/s00357-021-09389-2 

• 2020  Tomarchio S.D., Punzo A. and Bagnato L., Two new matrix-variate distributions with application in model-based clustering, Computational Statistics & Data Analysis, 152, 107050. DOI: https://doi.org/10.1016/j.csda.2020.107050 

• 2020 Tomarchio S.D. and Punzo A., Dichotomous unimodal compound models: Application to the distribution of insurance losses, Journal of Applied Statistics, 47 (13-15), 2328–2353. DOI: https://doi.org/10.1080/02664763.2020.1789076 

• 2019 Tomarchio S.D. and Punzo A., Modelling the loss given default distribution via a family of zero-and-one inflated mixture models, Journal of the Royal Statistical Society: Series A, 182(4): 1247–1266. DOI: https://doi.org/10.1111/rssa.12466 

B) Atti di Convegno 

• 2021 Tomarchio S.D., Gallaugher M.P.B., Punzo A. and McNicholas P.D. Finite mixtures of regressions with random covariates using multivariate skewed distributions, In: Book of short papers - SIS 2021.ISBN: 9788891927361. 

• 2020 Tomarchio S.D., Punzo A. and Bagnato L., A new mixture model for three-way data, In: Book of short papers - SIS 2020. ISBN: 9788891910776. 

• 2020 Tomarchio S.D., Ingrassia S. and Melnykov V., Parsimonious matrix normal mixtures: an application to university students indicators, In: Book of Abstracts, Third international conference on Data Science & Social Research, Bari, 10-11 Dicembre 2020. ISBN: 978-886629-051-3. 

-Presentation of papers at conferences or congress: 

A) In sessioni invitate 

• Model based-clustering via mixtures of two new matrix-variate distributions, “13th International Conference of the ERCIM WG on Computational and Methodological Statistics”, Virtual Conference, 19-21 Dicembre 2020. 

• Finite mixtures of matrix-variate regressions with random covariates, “16th Conference of the International Federation of Classification Societies”, Thessaloniki (Grecia), 26-29 Agosto 2019. 

• Clustering via new parsimonious mixtures of heavy tailed distributions, "13th Scientific Meeting of the Classification and Data Analysis Group", Firenze, Settembre 9-11, 2021 

B) In sessioni spontanee 

• Finite mixtures of regressions with random covariates using multivariate skewed distributions, “50th Scientific Meeting of the Italian Statistical Society”, Pisa(Italia), 21–25 Giugno 2021. 

• Parsimonious matrix normal mixtures: an application to university students indicators, “Third international conference on Data Science & Social Research”, Bari (Italia), 10–11 Dicembre 2020. 

• Clustering three-way data with a new mixture of Gaussian scale mixtures, “5th International Workshop: MBC 2 - Workshop on Model-Based Clustering and Classification”, Catania (Italia), 2, 4, 11, 18, 25 Settembre 2020. 

• The matrix-variate Gaussian Cluster-Weighted Model, “2019 Classification Society Annual Meeting”, MacEwan University, Edmonton (Canada), 19-22 Giugno 2019. 

• Modeling insurance losses via contaminated unimodal distributions, “49th Scientific Meeting of the Italian Statistical Society”, Università degli Studi di Palermo, Palermo (Italia), 20-22 Giugno, 2018.