Imperial College London
Portrait Picture

ProfessorRobertoTrotta

Faculty of Natural Sciences, Department of Physics

Visiting Professor
 
 
 
//

Contact

 

+44 (0)20 7594 7793r.trotta Website CV

 
 
//

Assistant

 

Mrs Sheila Ekudo +44 (0)20 7594 2086

 
//

Location

 

1009Blackett LaboratorySouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Feroz:2011:10.1007/JHEP06(2011)042,
author = {Feroz, F and Cranmer, K and Hobson, M and Ruiz, de Austri R and Trotta, R},
doi = {10.1007/JHEP06(2011)042},
journal = {Journal of High Energy Physics},
title = {Challenges of profile likelihood evaluation in multi-dimensional SUSY scans},
url = {http://dx.doi.org/10.1007/JHEP06(2011)042},
volume = {2011},
year = {2011}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Statistical inference of the fundamental parameters of supersymmetric theoriesis a challenging and active endeavor. Several sophisticated algorithms have been employedto this end. While Markov-Chain Monte Carlo (MCMC) and nested sampling techniquesare geared towards Bayesian inference, they have also been used to estimate frequentistconfidence intervals based on the profile likelihood ratio. We investigate the performanceand appropriate configuration of MultiNest, a nested sampling based algorithm, whenused for profile likelihood-based analyses both on toy models and on the parameter spaceof the Constrained MSSM. We find that while the standard configuration previously usedin the literarture is appropriate for an accurate reconstruction of the Bayesian posterior,the profile likelihood is poorly approximated. We identify a more appropriate MultiNestconfiguration for profile likelihood analyses, which gives an excellent exploration of theprofile likelihood (albeit at a larger computational cost), including the identification of theglobal maximum likelihood value. We conclude that with the appropriate configurationMultiNest is a suitable tool for profile likelihood studies, indicating previous claims tothe contrary are not well founded.
AU  - Feroz,F
AU  - Cranmer,K
AU  - Hobson,M
AU  - Ruiz,de Austri R
AU  - Trotta,R
DO  - 10.1007/JHEP06(2011)042
PY  - 2011///
SN  - 1126-6708
TI  - Challenges of profile likelihood evaluation in multi-dimensional SUSY scans
T2  - Journal of High Energy Physics
UR  - http://dx.doi.org/10.1007/JHEP06(2011)042
UR  - http://hdl.handle.net/10044/1/29733
VL  - 2011
ER  -
Download