首页> 外文OA文献 >Lazy ABC
【2h】

Lazy ABC

机译:懒惰的aBC

摘要

Approximate Bayesian computation (ABC) performs statistical inference forotherwise intractable probability models by accepting parameter proposals whencorresponding simulated datasets are sufficiently close to the observations.Producing the large quantity of simulations needed requires considerablecomputing time. However, it is often clear before a simulation ends that it isunpromising: it is likely to produce a poor match or require excessive time.This paper proposes lazy ABC, an ABC importance sampling algorithm which savestime by sometimes abandoning such simulations. This makes ABC more scalable toapplications where simulation is expensive. By using a random stopping rule andappropriate reweighting step, the target distribution is unchanged from that ofstandard ABC. Theory and practical methods to tune lazy ABC are presented andillustrated on a simple epidemic model example. They are also demonstrated onthe computationally demanding spatial extremes application of Erhardt and Smith(2012), producing efficiency gains, in terms of effective sample size per unitCPU time, of roughly 3 times for a 20 location dataset, and 8 times for 35locations.
机译:当相应的模拟数据集与观测值足够接近时,近似贝叶斯计算(ABC)通过接受参数建议对其他难处理的概率模型进行统计推断。产生大量所需的模拟需要大量的计算时间。但是,在模拟结束之前通常很清楚它是没有希望的:它可能会产生较差的匹配或需要过多的时间。本文提出了惰性ABC,这是一种ABC重要度采样算法,通过有时放弃这种模拟来节省时间。这使ABC更具可扩展性,可用于仿真成本很高的应用。通过使用随机停止规则和适当的加权步骤,目标分布与标准ABC相同。在一个简单的流行病模型实例上,提出并说明了调整懒惰ABC的理论和实用方法。它们还在Erhardt和Smith(2012)的计算要求极高的空间极端应用中得到了证明,就每单位CPU时间的有效样本大小而言,对于20个位置的数据集,效率提高了大约3倍,而对35个位置的数据集则提高了8倍。

著录项

  • 作者

    Prangle, Dennis;

  • 作者单位
  • 年度 2014
  • 总页数
  • 原文格式 PDF
  • 正文语种 {"code":"en","name":"English","id":9}
  • 中图分类

相似文献

  • 外文文献
  • 中文文献
  • 专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号