zbMATH — the first resource for mathematics

Bounding the effects of proxy variables on instrumental-variables coefficients. (English) Zbl 0638.62112
Instrumental-variables models are considered in which one of the observed variables is a proxy for some unobserved true variable. The smallest correlation between the proxy and the unobserved true variable is computed, such that it is guaranteed that a coefficient’s sign has not been reversed by substituting the proxy.
Reviewer: J.Wolters

62P20 Applications of statistics to economics
62J99 Linear inference, regression
Full Text: DOI
[1] Crocker, T.D.; Schulze, W.; Ben-Davie, S.; Kneese, A.V., Methods development for assessing air pollution control benefits, ()
[2] Kleeper, S.; Leamer, E.E., Consistent sets of estimates for regressions with errors in all variables, Econometrica, 52, 163-183, (1984) · Zbl 0526.90030
[3] Krasker, W.S.; Pratt, J.W., Bounding the effects of proxy variables on regression coefficients, Econometrica, 54, 641-655, (1986) · Zbl 0605.62140
[4] Zellner, Arnold, Constraints often overlooked in analyses of simultaneous equations models, Econometrica, 40, 849-853, (1972) · Zbl 0274.62078
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.