On Selection of Statistics for Approximate Bayesian Computing (or the Method of Simulated Moments)

By Creel Michael and Kristensen Dennis
Computational Statistics and Data Analysis (2015)

7 companion websites

Data-driven stabilizations of goodness-of-fit tests

Working Paper (2021)

Fernández-de-Marcos Alberto and García-Portugués Eduardo

R codes for JMVA-D-21-00195

Working Paper (2022)

Najarzadeh Dariush

Partial identification of non-linear peer effects models with missing data

Swiss Journal of Economics and Statistics (2022)

Madeira Carlos

Deterministic Sampling Based on Kullback-Leibler Divergence and its Applications

Working Paper (2022)

Wang Sumin and Sun Fasheng

Assessing the quality of self-reported debt information in household surveys

Journal of Survey Statistics and Methodology (2022)

Madeira Carlos

Empirical likelihood inference for longitudinal data with covariate measurement errors: an application to the LEAN study

Working Paper (2021)

Zhang Yuexia, Qin Guoyou, Zhu Zhongyi, and Zhang Jiajia

Principal componentsof spatial, rank and Kendalls tau-covariance matrices for the population, randomsamples, real and simulated data

Working Paper (2021)

Koch Inge and Naito Kanta

Testing the first-order separability hypothesis for spatio-temporal point patterns

Computational Statistics & Data Analysis (2021)

Ghorbani Mohammad, Vafaei Nafiseh, Dvořák Jiří, and Myllymäki Mari

Ensemble Sparse Estimation of Covariance Structure for Exploring Genetic Disease Data

Computational Statistics & Data Analysis (2021)

Kang Xiaoning and Wang Mingqiu

Abstract

A cross validation method for selection of statistics for Approximate Bayesian Computing, and for related estimation methods such as the Method of Simulated Moments, is presented. The method uses simulated annealing to minimize the cross validation criterion over a combinatorial search space that may contain an extremely large number of elements. A first simple example, for which optimal statistics (for point estimation) are known from theory, shows that the method is able to select these optimal statistics out of a large set of candidate statistics. A second example of selection of statistics for a stochastic volatility model illustrates the method in a more complex case. Code to replicate the results, or to use the method for other applications, is provided.

See paper

Suggested citation

Creel M. and Kristensen D. (2015) On Selection of Statistics for Approximate Bayesian Computing (or the Method of Simulated Moments). Computational Statistics and Data Analysis.

See paper