Graduated in Economics. He holds a Ph.D. in Statistics from Bocconi University in Milan (2011). Assistant professor of Statistics at University of Cagliari since 2012. He has been visinting researcher at Universitat Pompeu Fabra (Barcelona) and at the University of Rhode Island (USA). His main area of research is causal inference with particular focus on matching and weighting algorithms and their applications to epidemiological data. He is also interested in history of statistics and probability. He teaches statistics to undergraduate students in economics and business and gives an annual seminar on statistical methods to graduate students.
– Cannas M, Conversano C, Mola F and Sironi E: Variation in caesarean delivery rates across hospitals. A Bayesian semi-parametric approach. Journal of Applied Statistics, 44(12), pp. 2095–2107, (2017) [article link] [poster]
– Arpino B, Cannas M: Propensity score matching with clustered data. An application to the estimation of the impact of caesarean section on the Apgar score. Statistics in Medicine, 35(12), pp. 2074–2091, (2016) [article link] [Rpackage]
– Cannas M, Sironi E, Mola F: Estimating the effect of prenatal care on birth outcomes. Book of abstracts CLADAG 2015 (forthcoming).
– Cannas M, Sironi, E: Hospital Differences in rates of Caesarean Deliveries in Sardinian region: an Observational Study; Biostatistics, Epidemiology and Public Health, 11(4), (2014). [article link]
– Conversano C, Cannas M and Mola F: A recursive partitioning algorithm for balancing covariates in causal inference studies; Book of Abstracts of the 9th Scientific Meeting of the CLAssification and Data Analysis Group of the Italian Statistical Society (CLADAG); University of Modena, September 18-20 (2013) Editor: CLEUP. [article link]
– Cannas M, Sironi, E: Variation in rates of obstetric interventions across hospitals in Sardinia: a multilevel analysis; Electronic Journal of Applied Statistical Analysis: Decision Support Systems and Services Evaluation, 4(1), (2013). [article].
Talks and presentations
— Better together? Comparing and combining Coaresened Exact Matching and Propensity Score Matching (with Bruno Arpino)
7th Conference of the European Survey Research Association, Lisbon, Portugal, July 2017.
— Propensity score matching with clustered data. (with Bruno Arpino)
Invited talk at the 19th European Young Statisticians Meeting (EYSM), Charles University in Prague, Czech Republic, September 2015. [slides]
— A semi-parametric Bayesian model for clustering hospitals by similarity in patients’ outcome: a study of cesarean sections rates in Sardinia (with C Conversano, E Sironi and F Mola).
Invited talk at the 7th International Conference of the ERCIM WG on Computational and Methodological Statistics, University of Pisa, September 2014.
— Statistica (Corso di Laurea in Economia e Gestione Aziendale)
— Short courses (free-choice) on Statistical Methods in previous academic years
- Introduction to sampling methods (2015): syllabus
- Statistics for economic applications (2016): syllabus
- Introduction to causal inference (Ph.D. course 2017): syllabus
- Introduction to Bayesian Statistics (Ph.D course 2018): Phd2018
Ultima modifica novembre 2017