Problems with Instrumental Variables Estimation when the Correlation between the Instruments and the Endogenous Explanatory Variable is Weak

By J. Bound, David A. Jaeger, Regina S. Baker. 1995

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Causal inference by using invariant prediction: identification and confidence intervals (2015, methodology): “ecients is empty and we reject the whole model. If the instrument is weak (small or vanishing), the in uence of Ion Xis small and a potential problem occurs for the instrumental variables approach [Bound et al., 1995], as the weak dependence makes the identication of using two-stage least squares ha”
Counterfactual Prediction with Deep Instrumental Variables Networks (2016, methodology): “In the ﬁrst stage, using (15) to avoid overﬁt guards against the ‘weak instruments’ bias [Bound et al., 1995] that can occur when the instruments are only weakly correlated with the policy variable.”
Automating the Selection of Model-Implied Instrumental Variables (2004, background): “In addition to being nonzero, the R2s should be nontrivial to avoid the problems that accompany weak IVs (see Bound, Jaeger, and Baker 1995).”
Estimates of the economic return to schooling for 28 countries (2002, background): “In any case, we can test the validity of our instruments following Bound et al. (1995). These authors and others have shown that caution is needed in choosing suitable instruments.”
The Potential Impact of Large-Scale Voucher Programs (2000, background): “in the estimates of private school effects ( Bound, Jaeger, & Baker, 1995 ).”
Weak Identification in Fuzzy Regression Discontinuity Designs (2016, background): “…in 990 articles according to Google Scholar, with 372 of these articles explicitly considering FRD.) Around the same time, the seminal works of Bound, Jaeger, and Baker (1995) and Staiger and Stock (1997) made weak identification in an instrumental variables (IV) context an important…”
Non-Production Benefits of Education: Crime, Health, and Good Citizenship (2011, unknown): “Adams (2002) also uses quarter of birth as an instrument for schooling; however, a large literature on weak instruments suggests that this may produce IV estimates biased towards OLS estimates (see, e.g., Bound, Jaeger, and Baker (1995)).”
How Does Capital Affect Bank Performance During Financial Crises (2013, background): “If the instruments are weak (the correlation with the endogenous regressors is low), IV estimates are biased toward the OLS estimates in finite samples (e.g., Bound, Jaeger, and Baker 1995; Staiger and Stock 1997).”
State Higher Education Spending and the Tax Revolt (2006, methodology): “…of major crimes and on the crude death rate as instruments.7 The Hausman test indicated that there is not a systematic difference between the instrumental variables and ordinary least squares regression coefficients.8 We also performed a Bound, Jaeger, and Baker (1995) test for weak instruments.”
Testing Instrument Validity for LATE Identification Based on Inequality Moment Constraints (2015, methodology): “3 See Bound, Jaeger, and Baker (1995), who contest the validity of quarter of birth instruments and present evidence on seasonal patterns of births that are related to family income, physical and mental health, and school attendance rates (factors that may be correlated with potential wages).”