Three Alumni in the EMOS Master Thesis Competition
Public Evaluation and Decision Making Master’s former students, Chloé Lallemand, Régis Relland and Suzanne Scott are competing in the 2021 EMOS Master thesis competition. Every other year, this event seeks to highlight official statistics as a research topic and put forward the outstanding work of young talents within the EMOS international network.
ENSAI’s Master’s in Public Evaluation and Decision Making is the only European Master in Official Statistics (EMOS) labelled program in France. Alumni Chloé Lallemand, Régis Relland and Suzanne Scott have submitted their abstracts after having been pre-selected by their academic supervisors, Brigitte Gelein, François Coquet and Catherine Benjamin.
Selecting the best EMOS Master theses among 30 programs
The selection committee consists of the EMOS Board, which has 14 members from universities and national statistical institutes across Europe, European System of Central Banks (ESCB), European Statistical Advisory Committee (ESAC) and Eurostat. Winners will be notified in December 2020.
Up to five winners of the competition will present their Master theses in the New Techniques and Technologies for Statistics (NTTS), a Eurostat conference series, which will take place in Brussels in March 2021.
The laureates and their supervisors will receive an “Honors Diploma”. Theses will be published on the NTTS and EMOS websites.
In 2019, Noémie Morenillas, ENSAI class of 2018, was one of the winners.
Chloé Lallemand, Régis Relland and Suzanne Scott have written their Master’s theses after their final internships at Institut des Politiques Publiques and Insee.
Chloé Lallemand: “Assessment of causal impact of reforms of the French pension system on private sectors wage earners”
Chloé Lallemand is an economist at Institut des Politiques Publiques. She followed the first year of the Master’s at Université de Rennes 1 before joining the class of 2019 in the Master’s in Public Evaluation and Decision Making.
Her master thesis is the outcome of a six-month internship at the French Institut des Politiques Publiques (IPP), a scientific partnership between the Paris School of Economics (PSE) and the Centre for Research in Economics and Statistics (CREST – ENSAE), whose aim is to promote quantitative analysis and evaluation of public policy.
Chloé Lallemand worked on the assessment of the redistributive effects of the future reform of the French pension system. More precisely, she ran two different analyses of reforms of the system.
Firstly, she did an ex-ante evaluation of the potential future system in debate using a microsimulation model and administrative data. The goal of this part was to examine the redistributive effects of changes in the computation rules of pension benefits and to describe which type of workers would win or lost with the new system.
Secondly, she did an ex-post evaluation of the impact of a rise in the retirement age parameters on the retirement claiming choice. To perform this part, she assessed a former reform which has increased the minimum and full rate ages, using administrative data.
Régis Relland: “Custom zoning from grid data according to socio-demographic criteria: Comparison of geographical clustering methods”
Régis Relland is in charge of disseminating the population census data on the website of the French national statistical institute. Moreover, he is responsible for the administration of the Insee cartographic site (statistiques-locales.insee.fr).
After graduating from ENSAI in 2017, he started working at Insee. In parallel to his day job, he followed the Master’s in Public Evaluation and Decision Making at ENSAI, specializing in Methodology for Official Statistics.
In June 2019, Insee disseminated new grid data on its website. This data are presented as a great potential for territorial analysis, but so far they are under-used, especially internally at Insee. From a methodological point of view, it is difficult to bring out synthetic messages from such fine data. In particular, usual classification methods are poorly applied because they can lead to create clusters close as regards the phenomenon under study, but geographically very dispersed.
The objective was therefore to select, test and compare different clustering methods, each presenting a compromise between homogeneous and well separated clusters from the point of view of the attributes, and also clusters making sense geographically.
Within a team, Régis Relland tested different clustering methods (geographical extension of standard clustering methods, specialized clustering methods, neural networks …) and developed an application in R Shiny for statisticians.
Suzanne Scott: “Enhancing seasonal adjustment using X13-ARIMA-SEATS: the case of production in the French quarterly accounts”
Suzanne Scott is a researcher for Insee’s Economic Studies Department. She carries out studies on firms and on the labour market. She is currently working on the impact of the health crisis on firms’ cash flows and on their behaviors.
A civil servant student at ENSAI until 2020, she enrolled in the Master’s in Public Evaluation and Decision Making, specializing in Methodology for Official Statistics, right after. She will be graduating in 2021.
Suzanne Scott’s thesis studied the impact that a change in seasonal adjustment method from X12-ARIMA run in SAS to X13-ARIM-SEATS run in R would have on the quality of the French quarterly accounts. These accounts are central to short-term economic analysis which makes their reliability essential.
She began by comparing the theories behind these methods before analyzing how each one affected the adjustment’s quality and the way the accounts reacted to revisions in the data and to new data points. This thesis allowed her to discover the flaws of the current method used by Insee’s quarterly accounts division and thus – as part of their ongoing software change from SAS to R – to implement a more thorough seasonal and calendar adjustment process.
ENSAI’s Master’s in Public Evaluation and Decision Making
Public administrations and agencies need highly qualified Statisticians and Data Scientists to create, manage, and exploit local, national and even international databases. The objective of the Master’s is to respond to these increasing demands for data experts in the field of public policy evaluation and decision making.
Graduates are well-prepared to work in scientifically and technically demanding jobs in Statistical Engineering, Social and Economic IT systems, and IT management related to Official Statistics.
The Master’s in Public Evaluation and Decision Making includes three specializations: Methodology for Official Statistics, Statistics and Data Management and Statistical Studies.
Find out more about ENSAI’s Master’s in Public Evaluation and Decision Making.