- Spatial aspects of Civil War
- Counterinsurgency Practices
- Conflict Processess and sub-national political violence
- Reactive Violence
- Conflict Forecasting and Early Warning Signals
- Quantitative Methods
Explaining the variation in sub-national diffusion of civil conflict: reactive violence and military strategies
I use spatial quantitative methods to study the impact of civilians victimisation on horizontal escalation of civil conflicts. My research relies on large volumes on fine-grained event-data and geo-referenced spatial co-variates. I work on methodological problems as well focusing on the the creation of synthetic counter-factuals to better understand the causal linkages between conflict processes.
- “Diplomatic Correspondence and Conflict Anticipation” – Boussalis, Chadefaux, Decadri, Salvi – 2018 – Work in Progress
- “On the beaten path: Violence against civilians and simulated conflict along road networks.” - Salvi, Williamson, Draper – 2019 - Forthcoming book chapter in Computational Conflict Research - Deutschmann et al. - Springer Nature
- Non-presenting author at 76th Annual MPSA Conference - Chicago, US - 5-8 April 2018
- Presenting author at Conflict Research Society Annual Conference 2019: “Rethinking Conflict Research and Practice in a Post-Liberal World” - Brighton, UK - 8-10 September 2019
2016 - 2017 - Research Assistant for Dr. Thomas Chadefaux, Trinity College Dublin Chadefaux, T. (2017). Conflict forecasting and its limits. Data Science, 1(1-2), 7-17.
2015 - 2016 - Research Assistant for Prof. Raffaele Marchetti - LUISS Guido Carli “The Impact of Globalisation on Domestic Political Risk”
Should you be interested in reading them, please, feel free to contact me.
- MA Thesis (Law and Government of the EU, Chair of Quantitative Methods, 2016) – Supervisor: Dr. Lorenzo De Sio
“Explaining the variation in Political Risk: the role of globalization”
The paper – intended as a replication of Marchetti and Vitale (2014) focused on transnational variables as determinants for political violence. This involved the creation of an updated dataset with Political Risk and globalisation indicators for 193 Countries from 1990 to 2013. My hypotheses on the non-linearity of the relation were tested with a Mixed Model.
- MA Thesis (International Relations, Chair of Security Studies – 2015) – Supervisor: Dr. Raffaele Marchetti
“Quantitative Assessment of Conflict Risk: a latent class regression model”.
The study – intended as a replication of the GCRI methodology – aimed to create a quantitative model for conflict risk assessment able to encompass multiple dimensions and layers of conflict characterising the Post-Cold War era. In particular, I focused on the variation of intrastate and interstate conflict onsets. I made use of a subset of the GCRI dataset (1989-2013) together with UCDP/PRIO conflict data for 193 countries. The analysis was based on a Latent Class Regression model with concomitant variables, a model that is able to detect the presence of subpopulations within the overall pool of observations.
- BA Thesis (Political Science, Chair of Statistics, 2013) – Supervisor: Dr. Antonello Maruotti
“Markov-Switching Models for financial time-series”.
The project aimed to determine how correlations between variable rates may vary over time and how volatility may affect such correlations. It built on the work by Jan Bulla (2006), analysing the three major exchange rates (USD, JPY, GBP) in a time span of 13 years. In this view, I assembled a dataset of roughly 10,000 observations and three variables (exchange rate returns). The resulting distributions were highly lepktokurtic and required a notable effort in terms of data manipulation. The data had been modelled with a Hidden Markov Model using Matlab.
I had the chance to write and publish several policy papers during my internship as OSint Analyst at the Joint Research Centre of the European Commission.