Simulation from the Normal Distribution Truncated to an Interval in the Tail

Zdravko Botev; Pierre L'Ecuyer

10th EAI International Conference on Performance Evaluation Methodologies and Tools

Research Article

Simulation from the Normal Distribution Truncated to an Interval in the Tail

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.4108/eai.25-10-2016.2266879,
    author={Zdravko Botev and Pierre L'Ecuyer},
    title={Simulation from the Normal Distribution Truncated to an Interval in the Tail},
    proceedings={10th EAI International Conference on Performance Evaluation Methodologies and Tools},
    publisher={ACM},
    proceedings_a={VALUETOOLS},
    year={2017},
    month={5},
    keywords={truncated normal normal quantile random variate generation inversion exact sampling},
    doi={10.4108/eai.25-10-2016.2266879}
}

Zdravko Botev
Pierre L'Ecuyer
Year: 2017
Simulation from the Normal Distribution Truncated to an Interval in the Tail
VALUETOOLS
ACM
DOI: 10.4108/eai.25-10-2016.2266879

Zdravko Botev¹, Pierre L'Ecuyer²^,*

1: University of New South Wales
2: DIRO, Université de Montréal

*Contact email: lecuyer@iro.umontreal.ca

Abstract

We study and compare various methods to generate a random variate from the normal distribution truncated to some finite or semi-infinite interval, with special attention to the situation where the interval is far in the tail. This is required in particular for certain applications in Bayesian statistics, such as to perform exact posterior simulations for parameter inference, but could have many other applications as well. We distinguish the case in which inversion is warranted, and that in which a rejection method is also fine. The algorithms are implemented and available in Java, R, and MATLAB, and the software is freely available.

Keywords: truncated normal normal quantile random variate generation inversion exact sampling

Published: 2017-05-03
Publisher: ACM

: http://dx.doi.org/10.4108/eai.25-10-2016.2266879