Alessandra Mattei
Research Subjects
Alessandra Mattei has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Fabrizia Mealli
Department of Statistics, Computer Science, Applications, University of Florence, Florence, Italy
3 shared publications - Giulio Grossi
University of Florence
2 shared publications - Marco Mariani
IRPET Regional Institute for Economic Planning of Tuscany, Florence, Italy
2 shared publications - Annalisa Caloffi
Department of Economics and Management University of Florence Florence Italy
1 shared publication - Bianca Lucia De Stavola
Great Ormond Street Institute of Child Health, University College London, London WC1N 1EH, United Kingdom
1 shared publication - Björn Bornkamp
Novartis Pharma AG Basel Switzerland
1 shared publication
Latest Publications
- Open Access
- 20 August 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
Spatial vertical regression for spatial panel data: evaluating the effect of the florentine tramway’s first line on commercial vitality
Synthetic control methods are commonly used in panel data settings to evaluate the effect of an intervention. In many cases, the treated and control units correspond to spatial units such as regions or neighborhoods. Our approach addresses the challenge of understanding how an intervention applied at specific locations influences the surrounding area. Traditional synthetic control applications do not address how to define the effective area of impact, the extent of treatment propagation across space, and how effects vary with distance from the treatment sites. To address these challenges, we introduce spatial vertical regression (SVR) within the Bayesian paradigm. SVR predict the missing potential outcomes in varying proximities to the treatment sites, while accounting for the spatial structure inherent in the data. Rooted in the vertical regression framework of the synthetic control method, SVR employs a Gaussian process to ensure that the imputation of missing potential outcomes is spatially coherent, reflecting the expectation that nearby treated areas have similar relationships to control areas and experience similar outcomes. This approach is particularly pertinent to our study on the Florentine tramway’s first line construction. We study its influence on the local commercial landscape, focusing on how business prevalence varies at different distances from the tram stops.
- 20 August 2026
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
Open AccessSpatial vertical regression for spatial panel data: evaluating the effect of the florentine tramway’s first line on commercial vitality
Synthetic control methods are commonly used in panel data settings to evaluate the effect of an intervention. In many cases, the treated and control units correspond to spatial units such as regions or neighborhoods. Our approach addresses the challenge of understanding how an intervention applied at specific locations influences the surrounding area. Traditional synthetic control applications do not address how to define the effective area of impact, the extent of treatment propagation across space, and how effects vary with distance from the treatment sites. To address these challenges, we introduce spatial vertical regression (SVR) within the Bayesian paradigm. SVR predict the missing potential outcomes in varying proximities to the treatment sites, while accounting for the spatial structure inherent in the data. Rooted in the vertical regression framework of the synthetic control method, SVR employs a Gaussian process to ensure that the imputation of missing potential outcomes is spatially coherent, reflecting the expectation that nearby treated areas have similar relationships to control areas and experience similar outcomes. This approach is particularly pertinent to our study on the Florentine tramway’s first line construction. We study its influence on the local commercial landscape, focusing on how business prevalence varies at different distances from the tram stops.
- 23 February 2026
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 82 (2)
A First Course in Causal Inference
- 23 February 2026
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 82 (2)
A First Course in Causal Inference
- 13 November 2025
- Published by Wiley in Biometrical Journal
- Vol. 67 (6) , e70092
Evaluating Causal Effects on Time‐to‐Event Outcomes in an RCT in Oncology With Treatment Discontinuation
In clinical trials, patients may discontinue treatments prematurely, breaking the initial randomization. In our motivating study, a randomized controlled trial in oncology, patients assigned the investigational treatment may discontinue it due to adverse events. The ICH E9(R1) Addendum provides guidelines for handling such “intercurrent events.” The right strategy to adopt depends on the questions of interest. We propose adopting a principal stratum strategy and decomposing the overall intention‐to‐treat effect into principal causal effects for groups of patients defined by their potential discontinuation behaviour. We first show how to implement a principal stratum strategy to assess causal effects on a survival outcome in the presence of continuous‐time treatment discontinuation, its advantages, and the conclusions that can be drawn. Our strategy allows us to properly handle the time‐to‐event intermediate variable, which is not defined for patients who would not discontinue, and to account for the fact that the discontinuation time and the primary endpoint are subject to censoring. We employ a flexible model‐based Bayesian approach to tackle these complexities, providing easily interpretable results. We apply this Bayesian principal stratification framework to analyze synthetic data of the motivating oncology trial. Supported by a simulation study, we shed light on the role of covariates in this framework. Beyond making structural and parametric assumptions more credible, they lead to more precise inference. Also, they can be used to characterize patients' discontinuation behavior, which could help inform clinical practice and future protocols.
- 13 November 2025
- Published by Wiley in Biometrical Journal
- Vol. 67 (6) , e70092
Evaluating Causal Effects on Time‐to‐Event Outcomes in an RCT in Oncology With Treatment Discontinuation
In clinical trials, patients may discontinue treatments prematurely, breaking the initial randomization. In our motivating study, a randomized controlled trial in oncology, patients assigned the investigational treatment may discontinue it due to adverse events. The ICH E9(R1) Addendum provides guidelines for handling such “intercurrent events.” The right strategy to adopt depends on the questions of interest. We propose adopting a principal stratum strategy and decomposing the overall intention‐to‐treat effect into principal causal effects for groups of patients defined by their potential discontinuation behaviour. We first show how to implement a principal stratum strategy to assess causal effects on a survival outcome in the presence of continuous‐time treatment discontinuation, its advantages, and the conclusions that can be drawn. Our strategy allows us to properly handle the time‐to‐event intermediate variable, which is not defined for patients who would not discontinue, and to account for the fact that the discontinuation time and the primary endpoint are subject to censoring. We employ a flexible model‐based Bayesian approach to tackle these complexities, providing easily interpretable results. We apply this Bayesian principal stratification framework to analyze synthetic data of the motivating oncology trial. Supported by a simulation study, we shed light on the role of covariates in this framework. Beyond making structural and parametric assumptions more credible, they lead to more precise inference. Also, they can be used to characterize patients' discontinuation behavior, which could help inform clinical practice and future protocols.
- Open Access
- 6 February 2025
- Published by Elsevier in Econometrics and Statistics
Bayesian principal stratification with longitudinal data and truncation by death
- 6 February 2025
- Published by Elsevier in Econometrics and Statistics
Open AccessBayesian principal stratification with longitudinal data and truncation by death
- 1 October 2020
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 183 (4) , 1329-1332
Preface to the Papers on ‘Causal Inference from Non-experimental Studies: Challenges, Developments and Applications’
- 1 October 2020
- Published by Oxford University Press (OUP) in Journal of the Royal Statistical Society Series A: Statistics in Society
- Vol. 183 (4) , 1329-1332
Preface to the Papers on ‘Causal Inference from Non-experimental Studies: Challenges, Developments and Applications’
- Open Access
- 14 March 2020
- Published by Elsevier in Journal of Multivariate Analysis
- Vol. 178, 104609
Joint and marginal causal effects for binary non-independent outcomes
- 14 March 2020
- Published by Elsevier in Journal of Multivariate Analysis
- Vol. 178, 104609
Open AccessJoint and marginal causal effects for binary non-independent outcomes
- Open Access
- 5 September 2019
- Published by Elsevier in Papers in Regional Science
- Vol. 99 (1) , 25-54
What kinds of R&D consortia enhance SMEs productivity? A hierarchical Bayesian approach for the analysis of a regional innovation policy
- 5 September 2019
- Published by Elsevier in Papers in Regional Science
- Vol. 99 (1) , 25-54
Open AccessWhat kinds of R&D consortia enhance SMEs productivity? A hierarchical Bayesian approach for the analysis of a regional innovation policy