A global sensitivity test for evaluating statistical hypotheses with nonidentifiable models.

  title={A global sensitivity test for evaluating statistical hypotheses with nonidentifiable models.},
  author={David Todem and Jason Peter Fine and Lina Peng},
  volume={66 2},
We consider the problem of evaluating a statistical hypothesis when some model characteristics are nonidentifiable from observed data. Such a scenario is common in meta-analysis for assessing publication bias and in longitudinal studies for evaluating a covariate effect when dropouts are likely to be nonignorable. One possible approach to this problem is to fix a minimal set of sensitivity parameters conditional upon which hypothesized parameters are identifiable. Here, we extend this idea and… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.
6 Citations
24 References
Similar Papers


Publications referenced by this paper.
Showing 1-10 of 24 references

Identification and estimation of statistical functionals from incomplete data

  • J. L. Horowitz, C. F. Manski
  • Journal of Econometrics
  • 2006
2 Excerpts

Ignorance and uncertainty regions as inferential tools in a sensitivity analysis

  • S. Vansteelandt, E.J.T. Goetghebeur, M. G. Kenward, G. Molenberghs
  • Statistica Sinica
  • 2006
2 Excerpts

Observational Studies. New York: Springer

  • P. R. Rosenbaum
  • 2002
1 Excerpt

Local sensitivity approximations for selectivity bias

  • J. Copas, S. Eguchi
  • Journal of the Royal Statistical Society, Series…
  • 2001
2 Excerpts

Similar Papers

Loading similar papers…