Table 7:

OLS, IV and Upper Bounds Estimates for the Effect of BMI on Depression

	Mean (1)	OLS (2)	IV (3)	IV(1) (4)	UB of β under (A4) (5)	UB of β (A4 & A5) (6)
Panel A: Add health
All	0.15	0.002*** (.001)	0.003 (.003)	0.002 (.001)	0.002 [.004]	0.002 [.004]
Women	0.17	0.004*** (.001)	0.004 (.004)	0.004** (.002)	0.004 [.006]	0.004 [.0007]
Men	0.12	−0.002 (.0010)	−0.001 (.004)	−0.002 (.002)	−0.002 [.001]	−0.001 [.002]
Panel B: HRS
All	0.12	0.003*** (.001)	0.007*** (.003)	0.002** (.001)	0.003 [.005]	0.002 [.005]
Women	0.14	0.004*** (.001)	0.007** (.004)	0.003* (.001)	0.004 [.006]	.003 [.006]
Men	0.09	0.003*** (.001)	0.007* (.004)	0.002 (.002)	0.003 [.005]	0.001 [.006]

Notes: Add Health regressions control for age, age squared, gender, birth order, mother’s education, picture vocabulary score, PGSs for depression and education, and the first 20 ancestry-specific principal components of the genetic data. HRS regressions control for age, age squared, gender, mother’s education, PGSs for depression and education, and the first 10 ancestry-specific principal components of the genetic data. Heteroscedasticity-robust standard errors in (.)

^***significant at 1%

^**significant at 5%

^*significant at 10%.

The Nevo & Rosen (2012) approach is implemented using the imperfectiv command in Stata. The upper endpoint of the 95% confidence interval on the bounded parameter is given in [.].