Hostname: page-component-586b7cd67f-gb8f7 Total loading time: 0 Render date: 2024-11-25T08:56:49.291Z Has data issue: false hasContentIssue false

A careful consideration of CLARIFY: simulation-induced bias in point estimates of quantities of interest – CORRIGENDUM

Published online by Cambridge University Press:  07 July 2023

Carlisle Rainey*
Affiliation:
Department of Political Science, Florida State University, Tallahassee, USA ([email protected]).
Rights & Permissions [Opens in a new window]

Abstract

Type
Corrigendum
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of the European Political Science Association

The author would like to amend an error in the above article as per the following:

In Rainey (Reference Rainey2023), I evaluate the point estimate suggested by King, Tomz, and Wittenberg (Reference King, Tomz and Wittenberg2000). They suggest that researchers “[a]verage the simulated values to obtain a point estimate” (p. 351). I show that this approach creates “simulation-induced bias” and recommend that researchers directly transform the maximum likelihood estimates instead.

In listing examples of software that implement the two approaches, I regrettably describe a new R package clarify incorrectly (Greifer et al. Reference Greifer, Worthington, Iacus and King2023). In footnote 1 and the second paragraph of the conclusion, I wrongly cite the new clarify for R as an example of software that uses the average of simulations as the point estimate.

Predecessors CLARIFY for Stata (version 2.0; Tomz et al., Reference Tomz, Wittenberg and King2003,) and Zelig (version 5.1.7; Imai et al., Reference Imai, King and Lau2008; Choirat et al., Reference Choirat, Honaker, Imai, King and Lau2018) use the average of simulations to compute a point estimate. clarify for R was released January 25, 2023 (version 0.1.0) as a replacement for the now-deprecated R package Zelig, and the new R package clarify “directly transform[s] maximum likelihood estimates of coefficients to obtain maximum likelihood estimates of the quantities of interest” as Rainey (Reference Rainey2023) suggests.

References

Choirat, C, Honaker, J, Imai, K, King, G and Lau, O (2018) Zelig: Everyone's Statistical Software. Version 5.1.7 (December 12, 2020). http://zeligproject.org/Google Scholar
Greifer, N, Worthington, S, Iacus, S and King, G (2023) clarify: Simulation-Based Inference for Regression Models. Version 0.1.0 (January 25, 2023). https://iqss.github.io/clarify/Google Scholar
Imai, K, King, G and Lau, O (2008) Toward a common framework for statistical analysis and development. Journal of Computational and Graphical Statistics 17, 892913.CrossRefGoogle Scholar
King, G, Tomz, M and Wittenberg, J (2000) Making the most of statistical analyses: improving interpretation and presentation. American Journal of Political Science 44, 341355.CrossRefGoogle Scholar
Tomz, M, Wittenberg, J and King, G (2003) Clarify: software for interpreting and presenting statistical results. Journal of Statistical Software 8, 130CrossRefGoogle Scholar
Rainey, C. (2023). A careful consideration of CLARIFY: simulation-induced bias in point estimates of quantities of interest. Political Science Research and Methods, 110. doi:10.1017/psrm.2023.8Google Scholar