论文信息 - Very Good Importance Sampling

Very Good Importance Sampling

Importance weighting is a convenient general way to adjust for draws from the wrong distribution, but the resulting ratio estimate can be noisy when the importance weights have a heavy right tail, as routinely occurs when there are aspects of the target distribution not well captured by the approximating distribution. More stable estimates can be obtained by truncating the importance ratios. Here we present a new method for stabilizing importance weights using generalized Pareto distribution t to

Aki Vehtari | A. Gelman | Aki Vehtari | Andrew E. Gelman

[1] Aki Vehtari,et al. Understanding predictive information criteria for Bayesian models , 2013, Statistics and Computing.

[2] M. Pitt,et al. On the existence of moments for high dimensional importance sampling , 2013, 1307.7975.

[3] Jin Zhang,et al. A New and Efficient Estimation Method for the Generalized Pareto Distribution , 2009, Technometrics.

[4] Drew D. Creal,et al. Testing the assumptions behind importance sampling , 2009 .

[5] S. MacEachern,et al. Case-deletion importance sampling estimators: Central limit theorems and related results , 2008, 0807.0725.

[6] E. Ionides. Truncated Importance Sampling , 2008 .

[7] Louis H. Y. Chen,et al. Normal approximation under local dependence , 2004, math/0410104.

[8] Christian P. Robert,et al. Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[9] R. Runde,et al. On the existence of moments , 1998 .

[10] Mario Peruggia,et al. On the variability of case-deletion importance sampling weights in the Bayesian linear model , 1997 .

[11] Hong Chang,et al. Model Determination Using Predictive Distributions with Implementation via Sampling-Based Methods , 1992 .

[12] J. Geweke,et al. Bayesian Inference in Econometric Models Using Monte Carlo Integration , 1989 .

[13] J. Pickands. Statistical Inference Using Extreme Order Statistics , 1975 .