• Aucun résultat trouvé

“Stress Testing” (joint program with BNP Paribas)

N/A
N/A
Protected

Academic year: 2022

Partager " “Stress Testing” (joint program with BNP Paribas) "

Copied!
2
0
0

Texte intégral

(1)

Lecturer Opportunity at Ecole Polytechnique, France

In Stochastic Simulation, Uncertainty Quantification, Statistical modeling

Chaire

“Stress Testing” (joint program with BNP Paribas)

Location. École Polytechnique is a French public institution of higher education and research in Palaiseau, in the southwest of Paris. It was established in 1794 by the mathematician Gaspard Monge during the French Revolution. It is one of the most prestigious and selective French grandes écoles. Polytechnique has several research laboratories operating in various scientific fields (physics, mathematics, computer science, economics, chemistry, biology, etc.).

Research program. The Chaire “Stress Testing” is a specific research program between Ecole Polytechnique and BNP Paribas, and is hosted at Polytechnique by the Center of Applied Mathematics https://portail.polytechnique.edu/cmap/en

This research project is part of an in-depth reflection on the increasingly sophisticated issues surrounding stress tests (under the impulse of the upcoming European Banking regulation).

Simulation of extreme adverse scenarios is an important topic to better understand which critical configurations can lead to financial and systemic crises. These scenarios may depend on complex phenomena, for which we partially lack information, making the modeling incomplete and uncertain. Last, the data are multivariate and reflect the dependency between driving variables.

From the above observations, different lines of research are considered:

1) the generation of stress test and meta-modeling scenarios using machine learning 2) the quantification of uncertainties in risk metrics

3) modeling and estimation of multidimensional dependencies

Keywords: Bayesian Networks, copulas, dependent data, Deep Learning, Gaussian processes, machine learning, Markov Chain Monte Carlo, meta-modeling, multivariate statistics, rare event simulation, risk metrics, splitting methods, stochastic algorithms, stochastic and Bayesian optimization, uncertainty propagation

Candidate profile:

• A PhD in Probability and Statistics, or equivalent

• A proven track record in quality research, as evidenced by research publications in the top scientific journals and conferences

• Solid working knowledge of some of the topics listed above (see keywords), both in research and teaching

• Solid working knowledge in the development of numerical methods or data analysis (with Python, R)

• An understanding of financial risks is desirable

• Good communication and management skills, allowing to be involved in the scientific life of the Chaire

Position: 3 years, extension is possible.

The teaching load will be about 70 hours/year.

Net salary: about 2800€/month depending on the experience

(2)

How to Apply: Your application should include a Cover Letter, Resume, publications,

description of teaching experience. With your application, we ask that you briefly outline your experience against the selection criteria in the position description.

Contact for application: emmanuel.gobet@polytechnique.edu

Références

Documents relatifs

Taken together, our data show that the 2G4 mAb can reliably discriminate Vangl2 from Vangl1 in human and mouse protein extracts, and that homozygous Lp mutant mice express the

A methodological caveat is in order here. As ex- plained in §2.1, the acoustic model takes context into account. It is theoretically possible that the neural network learnt

Mais, c’est l’édification de la ville qui, deux siècles plus tard, révèle la fragilité du milieu.. En effet, la fin du XVIII e siècle marque une rupture,

Keywords Bayesian statistics · Density dependence · Distance sampling · External covariates · Hierarchical modeling · Line transect · Mark-recapture · Random effects · Reversible

Monte Carlo methods, Markov Chain Monte Carlo, Sequential Monte Carlo, Bayesian modeling, Gaussian Processes, environmental statistics,

Table 3 presents the results for these cases with an a priori spectral index equal to the real value α = 1.1 (case IIIa), and without any a priori value on any parameter (case IIIb)

The present article (Mackay-Sim et al., 2008) reports on the effect of this approach in six chronic complete (ASIA A) spinal cord injury subjects (phase I/IIa design) with

Key words: Curve Resolution, Factor analysis, Bayesian estimation, non-negativity, statistical independence, sparsity, Gamma distribution, Markov Chain Monte Carlo (MCMC)..