Prof. Elena Stanghellini


Click here for my contact details and web page.

At 360 Port Avenue, with Elizabeth, John and Marco

Research interests

All my research has been devoted to Graphical Markov models. The class of Graphical Markov models is so vast that includes many existing models, like Structural Equation Models, log-linear models, and so on, that I never get tired in exploring them from the theoretical point of view and in using them in applied data analysis.


From the theoretical point of view I have been attracted by issues of identification and distortion induced by unobserved factors. For that I have also received a grant from the American Institute of Mathematics AIM ('SQuaRE' grant: 2011-2013).  Papers that address the identification issues are e.g. 4, 5 , 8 of the paper list below. Papers that address issue of distortion are e.g. 2, 6, 11, 13. In many applied research I have been using Graphical Models with latent variables: see papers 1, 2, 7, 12 or the recent book chapter 1 on capture-recapture estimation.


Currently, I am a member of SIMSAM,   Interdisciplinary research group on early life exposure and health: A registry-based life-course  perspective with modern epidemiological and statistical approches, at Umeå University.


Further Research Interests include Credit Scoring see e.g. paper 3 or 9, and book 4.

Books or chapters in books

1. STANGHELLINI E., RANALLI M.G. (2017). “Population size estimation using a categorical latent variable”. In Boehning D., Bunge J., van der Heijden P.G. Capture-Recapture Methods for the Social and Medical Sciences. Chapmand and Hall. ISBN: 9781498745314.


2. MEALLI F., PACINI B., STANGHELLINI E. (2014). “Identification of principal causal effects using secondar outcomes”. In Carpita M., Brentari E.,

Qannari E. Advances in Latent Variables. Studies in Theoretical and Applied Statistics. Springer-Verlag, doi: 10.1007/10104_2014_15.


3. GOTTARD A., STANGHELLINI E., CAPOBIANCO R. (2013).  “Semicontinuous regression models with Skew distributions”. In: Grigoletto M., Lisi F., Petrone S.. Complex Models and Computational Methods in Statistics . Springer Verlag, ISBN: 9788847028708, doi: 10.1007/978-88-470-2871-5-12.


4. STANGHELLINI E. (2009). Introduzione ai metodi statistici per il Credit Scoring. Springer-Verlag. (in Italian).

Papers (from 2004)

  1. MEALLI F., PACINI B., STANGHELLINI E. (2016). “Identification of principal causal effects using secondary outcomes in concentration graphs”. Journal of Educational and Behvioural Statistics, doi:10.3102/1076998616646199. Vol. 41, pp. 463-480.
  2. DORETTI M., GENELETTI S., STANGHELLINI E. (2016). “Tackling non.ignorable dropout in the presence of time varying confounding”. Journal of the Royal Statistical Society, Series C, doi: 10.1111/rssc.12154. Vol. 65, pp. 775-795.
  3. PIERRI F., STANGHELLINI E., BISTONI N. (2016). “Risk analysis and retrospective unbalanced data”. REVSTAT, 14, Vol. 2, pp. 157-169.
  4. ALLMAN E.S., RHODES J. A., STANGHELLINI E., VALTORTA M. (2014). “Parameter identifiability of Discrete Bayesian Networks with Hidden Variables.”Journal of Causal Inference, doi: 10.1515/jci-2014-0021.
  5. STANGHELLINI E., PAKPAHAN E. (2014). “Identification of Causal Effects in Linear Models: beyond Instrumental Variables”. TEST, doi: 10.1007/s11749-014-0421-3.
  6. GENBÄCK, M., STANGHELLINI E., DE LUNA X. (2014). “Uncertainty Intervals for regression parameters with non-ignorable missingness uin the outcome. Statistical Papers, doi: 10.1007/s00362-014-0610-x.
  7. NICOLOSI M., GRASSI S.,STANGHELLINI E. (2014). “Item Response Models to Measure Corporate Social Responsibility”. Applied Financial Economics, doi: 10.1080/09603107.2014.925070.
  8. STANGHELLINI E., VANTAGGI B. (2013). On the identification of discrete concentration graph models with one hidden binary variable ”. Bernoulli, 19, Number 5A, 1920-1937, doi:  10.3150/12-BEJ435.
  9. PIERRI F., BURCHI A. STANGHELLINI E. (2013). “La capacità predittiva degli indicatori di bilancio: un metodo per le piccolee medie imprese”, Piccola Impresa-Small Business, vol. 1, p. 85-108.
  10. STANGHELLINI E. (2012). Contribution to the Discussion to the paper di Wermuth N. and Sadeghi K. “Sequences of regressions and their independences”, TEST, vol. 21, p. 265-267, doi: 10.1007/s11749-012-0287-1.

11.  STINGO F.C., STANGHELLINI E., CAPOBIANCO R. (2011). “On the estimation of a binary response model in a selected population”. Journal of Statistical Planning and Inference, 141, pp. 3293-3303, doi:10.1016/j.jspi.2011.04.014. Click here for the R software to estimate the ESP model.

12.  HUTTON J.L., STANGHELLINI E. (2011). “Modelling bounded health scores with censored skew-normal distributions”. Statistics in Medicine, 30, pp. 368-376, doi: 10.1002/sim.4104.

13.  FORCINA A., GIOVAGNOLI A., STANGHELLINI E. (2011). “Non-compliance in surgical patients with herniated lumbar discs: an application of an extended latent class model as a selection model”. Statistical Modelling, 11(4), pp. 311-324, doi: 10.1177/1471082X1001100402.

14.  FALOCCI N., PANICCIA’ R.,  STANGHELLINI E. (2009). “Regression modelling of the flows  in an input-output table with accounting constraints”. Statistical Papers, 50, pp. 373-382.

15.   GIORGI ROSSI. P., MANTOVANI J., FERRONI J., FORCINA A., STANGHELLINI E., CURTALE F., BORGIA P. (2009). “Bacterial meningitis surveillance in Lazio, Italy: a system integrating laboratory specificity to hospitalization database sensitivity”. BMC Infectious Diseases.  

16.  MARCHETTI, G.M., STANGHELLINI, E. (2008), " A note on distortions induced by truncation, with application to linear regression systems". Statistics & Probability Letters, 78, pp.824-829.

17.  STANGHELLINI, E. (2006), " On statistical issues raised by the New Capital Accord ". Statistica Applicata, 18, 2, pp. 389-405.

18.  STANGHELLINI, E. , WERMUTH, N. (2005)," On the identification of path analysis models with one hidden variable". Biometrika, 92, 2, pp. 337-350.

19.  STANGHELLINI E., VAN DER HEIJDEN P.G.M. (2004), "A multiple-record systems estimation method that takes observed and unobserved heterogeneity into account". Biometrics, 60, pp. 510-516.

20.  STANGHELLINI, E. (2004), "Instrumental variables in Gaussian directed acyclic graph models with an unobserved confounder". Environmetrics, 15, pp. 463-469.