Post-Doc position FDS-IRPET: Statistical information systems for public investment projects

A post-doc position is available in the Department of Statistics, Computer Science, Applications “G. Parenti” at the University of Florence, as part of a reasearch agreement between the Florence Center for Data Science and the Regional Institute for Economic Planning of Tuscany (IRPET-Istituto Regionale Programmazione Economica della Toscana), and under the supervision of Prof. Cristina Martelli.
The ideal candidate should have some knowledge/interest in (statistical) information systems, statistical learning, machine learning, public economics.

Design, management and use of statistical information systems for the analysis and forecasting of public investment projects issues and inefficiencies

Research topic
Private and public Investments are strategical factors for growth and competitiveness: public sector investment is so at the heart of European and national economic policy, but the signs of recovery are still insufficient. This project intends to focus on the causes that, up to now, have prevented or limited new infrastructure projects and the maintenance and enrichment of public assets, with an emphasis to highlighting critical issues and inefficiencies. This project has the objective of designing and develop a knowledge system based on the integrated reuse of administrative data bases and available and pertinent statistical source.
The system will use a variety of sources, produced, among others by Agenzia di Coesione Territoriale, Tuscan Region, Opencoesione, SIOPE, SIMOG-ITACA
These sources are usually generated along the administrative processes of expenditure management, construction and investments’ results control: a multiplicity of different protocols and procedures in the responsibility of different institutions, with own specific administrative languages and semantics.
Due to the poor quality of the information system that it is not statistically reusable by design, these sources are often used only with a “big data” and descriptive approach, as they are not able to support the analysis with interpretable and causal models.
The knowledge system, object of this project, will take advantage of an integrated use of new technologies: relational and non relational data bases, semantic approaches and AI.
The information system will support statistical and machine learning analyses for predicting time of realization of public investments and the construction and use of data-driven indicators for anomaly detection.

Details can be found here.

Deadline for application: November 22nd 2019
Interview: December 4th 2019