README.md

In seschaub/robomit: Robustness Checks for Omitted Variable Bias

robomit

Summary

In nonexperimental work, omitted variables bias can cause flawed research conclusions. robomit implements the recent framework by Oster (2019) in R, which assesses the potential severity of the omitted variable bias for the research conclusion. For this assessment, we can estimate i) the bias-adjusted treatment effect or correlation and ii) the degree of selection on unobservables relative to observables that would be necessary to eliminate the result based on the Oster framework, using standard regression output. robomit implements an easy-to-use estimation of both of these variables. Additionally, robomit includes sensitivity analyses of these variables and their visualization. Find introduction to the package here.

Install package

# from CRAN
install.packages("robomit")

References

Oster. Unobservable selection and coefficient stability: Theory and evidence. Journal of Business & Economic Statistics, 37(2):187–204, 2019. URL https://doi.org/10.1080/07350015.2016.1227711.

seschaub/robomit documentation built on July 1, 2022, 7:49 p.m.