Roberto Di Mari

Assistant Professor
Department of Economics and Business
University of Catania 
Corso Italia, 55, 95128, Catania, Italy.
Tel: +(39) 095753764
Email: roberto.dimari@unict.it
ORCID: https://orcid.org/0000-0001-5498-009X
Research Gate: https://www.researchgate.net/profile/Roberto_Di_Mari
Google scholar: Roberto Di Mari

Curriculum Vitae (last update May 16, 2018): Download CV

Education

• Ph.D. (with distinction) in Economics and Finance, 2017, University of Rome Tor Vergata, Italy; Ph.D.
  thesis on Finite Mixture Models, supervised by Prof. R. Rocci (University of Rome Tor Vergata) and
  co-supervised by Prof. J. K. Vermunt (Tilburg University). Attended the one-year graduate program at
  the Einaudi Institute for Economics and Finance (EIEF, Bank of Italy, 2013).
• Visiting PhD student, 2015, at the Department of Methodology and Statistics of Tilburg University
  (Tilburg, Netherlands).
• Master of Science in Economics, 2011-2013, University of Rome Tor Vergata, Rome, Italy; final dissertation
  in Statistics, on Finite Mixtures of Linear Models: Numerical Evidences and Application to SHIW
  Data, supervised by Prof. R.Rocci (University of Rome Tor Vergata) with final grade: 110/110.
• Bachelor Degree in Economics, 2011, University of Catania, Italy; final dissertation in Statistics, Peer
  effects on academic outcome supervised by Prof. S.Ingrassia (University of Catania).
•  Erasmus Exchange Student, 2009 - 2010, University of Lille 1, France

Research Interests
Model-Based Clustering; Finite Mixture models; Latent Variable models; Latent Class analysis; Latent (Hidden) Markov models.

• Catania L., and Di Mari R. (2018). Hierarchical Hidden Markov Models for Multivariate Integer-valued Time-series. Journal of Econometrics (forthcoming).
• Di Mari R., Bakk Z., and Punzo A. (2019). A random-covariate approach for distal outcome prediction with latent class analysis. Structural Equation Modeling: A Multidisciplinary Journal (forthcoming).
• Di Mari R., Rocci R., and Gattone S.A. (2019). Scale-constrained approaches for maximum likelihood estimation and model selection of clusterwise linear regression models. Statistical Methods and Applications (forthcoming).
• Di Mari R. and Bakk Z. (2018). Mostly harmless direct effects: a comparison of different latent Markov modeling approaches. Structural Equation Modeling: A Multidisciplinary Journal, 25(3), 467-483.
• Di Mari R., Rocci R., and Gattone S.A. (2017). Clusterwise linear regression modeling with soft scale constraints. International Journal of Approximate Reasoning, 91, 160-178.
• Rocci R., Gattone S.A., and Di Mari R (2017). A data driven equivariant approach to constrained Gaussian mixture modeling Advances in Data Analysis and Classification, (forthcoming).
• Di Mari R. Rocci R., and Gattone S.A. (2017). Finite mixture of linear regression model: an adaptive constrained approach to maximum likelihood estimation. In: "Ferraro M. et al (Eds.), Soft Methods for Data Science. Advances in Intelligent System and Computing", vol. 456, Springer, Switzerland.
• Di Mari R., Oberski D.L., and Vermunt J.K. (2016). Bias-adjusted three step latent Markov modeling with covariates. Structural Equation Modeling: A Multidisciplinary Journal, 23(5), 649-660.

 
Papers submitted to journals

• Catania L., Di Mari R. and Santucci de Magistris (2019). Dynamic Discrete Mixtures for High Frequency Prices.
• Di Mari R., and Maruotti, A. (2019). Time-varying measurement error in generalized linear models for longitudinal data: a two-step latent Markov approach for the analysis of the Chinese Longitudinal Healthy Longevity Survey.