New insights into Approximate Bayesian Computation

نویسندگان

  • Gérard Biau
  • Frédéric Cérou
  • Arnaud Guyader
چکیده

Approximate Bayesian Computation (abc for short) is a family of computational techniques which offer an almost automated solution in situations where evaluation of the posterior likelihood is computationally prohibitive, or whenever suitable likelihoods are not available. In the present paper, we analyze the procedure from the point of view of k-nearest neighbor theory and explore the statistical properties of its outputs. We discuss in particular some asymptotic features of the genuine conditional density estimate associated with abc, which is an interesting hybrid between a k-nearest neighbor and a kernel method. Index Terms — Approximate Bayesian Computation, Nonparametric estimation, Conditional density estimation, Nearest neighbor methods, Mathematical statistics. 2010 Mathematics Subject Classification: 62C10, 62F15, 62G20. Corresponding author. Research partially supported by the French National Research Agency (grant ANR09-BLAN-0051-02 “CLARA”) and by the Institut universitaire de France. Research carried out within the INRIA project “CLASSIC” hosted by Ecole Normale Supérieure and CNRS. 1 Résumé Le terme anglais “Approximate Bayesian Computation” (abc en abrégé) désigne une famille de techniques bayésiennes ayant pour objet la simulation selon une loi de probabilité lorsque la vraisemblance a posteriori n’est pas disponible ou s’avère impossible à évaluer numériquement. Dans le présent article, nous envisageons cette procédure du point de vue de la théorie des k-plus proches voisins, en nous attachant plus particulièrement à examiner les propriétés statistiques des sorties de l’algorithme. Cela nous conduit à analyser le comportement asymptotique d’un estimateur de la densité conditionnelle naturellement associé à abc, utilisé en pratique et possédant à la fois les caractéristiques d’un estimateur des k-plus proches voisins et celles d’une méthode à noyau. Mots-clés —Approximate Bayesian Computation, Estimation non paramétrique, Estimation de la densité conditionnelle, Méthodes de plus proches voisins, Statistique mathématique. Classification par Sujets Mathématiques 2010 : 62C10, 62F15, 62G20.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Ltter CodABC: A Computational Framework to Coestimate Recombination, Substitution, and Molecular Adaptation Rates by Approximate Bayesian Computation

The estimation of substitution and recombination rates can provide important insights into the molecular evolution of protein-coding sequences. Here, we present a new computational framework, called “CodABC,” to jointly estimate recombination, substitution and synonymous and nonsynonymous rates from coding data. CodABC uses approximate Bayesian computation with and without regression adjustment...

متن کامل

CodABC: A Computational Framework to Coestimate Recombination, Substitution, and Molecular Adaptation Rates by Approximate Bayesian Computation

The estimation of substitution and recombination rates can provide important insights into the molecular evolution of protein-coding sequences. Here, we present a new computational framework, called "CodABC," to jointly estimate recombination, substitution and synonymous and nonsynonymous rates from coding data. CodABC uses approximate Bayesian computation with and without regression adjustment...

متن کامل

Fundamentals and Recent Developments in Approximate Bayesian Computation

Bayesian inference plays an important role in phylogenetics, evolutionary biology, and in many other branches of science. It provides a principled framework for dealing with uncertainty and quantifying how it changes in the light of new evidence. For many complex models and inference problems, however, only approximate quantitative answers are obtainable. Approximate Bayesian computation (ABC) ...

متن کامل

Model Selection in Historical Research Using Approximate Bayesian Computation

FORMAL MODELS AND HISTORY Computational models are increasingly being used to study historical dynamics. This new trend, which could be named Model-Based History, makes use of recently published datasets and innovative quantitative methods to improve our understanding of past societies based on their written sources. The extensive use of formal models allows historians to re-evaluate hypotheses...

متن کامل

Speech Enhancement Using Gaussian Mixture Models, Explicit Bayesian Estimation and Wiener Filtering

Gaussian Mixture Models (GMMs) of power spectral densities of speech and noise are used with explicit Bayesian estimations in Wiener filtering of noisy speech. No assumption is made on the nature or stationarity of the noise. No voice activity detection (VAD) or any other means is employed to estimate the input SNR. The GMM mean vectors are used to form sets of over-determined system of equatio...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2012