@article{fdi:010070149,
  title = {{A} comparison of two {B}ayesian approaches for uncertainty quantification},
  author = {{M}ara, {T}.{A}. and {D}elay, {F}. and {L}ehmann, {F}. and {Y}ounes, {A}nis},
  editor = {},
  language = {{ENG}},
  abstract = {{S}tatistical calibration of model parameters conditioned on observations is performed in a {B}ayesian framework by evaluating the joint posterior probability density function (pdf) of the parameters. {T}he posterior pdf is very often inferred by sampling the parameters with {M}arkov {C}hain {M}onte {C}arlo ({MCMC}) algorithms. {R}ecently, an alternative technique to calculate the so-called {M}aximal {C}onditional {P}osterior {D}istribution ({MCPD}) appeared. {T}his technique infers the individual probability distribution of a given parameter under the condition that the other parameters of the model are optimal. {W}hereas the {MCMC} approach samples probable draws of the parameters, the {MCPD} samples the most probable draws when one of the parameters is set at various prescribed values. {I}n this study, the results of a user-friendly {MCMC} sampler called {DREAM}((zs)) and those of the {MCPD} sampler are compared. {T}he differences between the two approaches are highlighted before running a comparison inferring two analytical distributions with collinearity and multimodality. {T}hen, the performances of both samplers are compared on an artificial multistep outflow experiment from which the soil hydraulic parameters are inferred. {T}he results show that parameter and predictive uncertainties can be accurately assessed with both the {MCMC} and {MCPD} approaches.},
  keywords = {},
  booktitle = {},
  journal = {{E}nvironmental {M}odelling and {S}ofware},
  volume = {82},
  numero = {},
  pages = {21--30},
  ISSN = {1364-8152},
  year = {2016},
  DOI = {10.1016/j.envsoft.2016.04.010},
  URL = {https://www.documentation.ird.fr/hor/fdi:010070149},
}