In this article, we examine several techniques that allow for the construction of bounds estimates based on instrumental variables, even when the instruments are not valid. We introduce the plausexog and imperfectiv commands, which implement methods described by Conley, Hansen, and Rossi (2012, Review of Economics and Statistics 94: 260–272) and Nevo and Rosen (2012b, Review of Economics and Statistics 94: 659–671). We examine the performance of these bounds under a range of circumstances, which leads to several practical results related to the informativeness of the bounds in different situations.
AngristJ. D., and EvansW. N.1998. Children and their parents’ labor supply: Evidence from exogenous variation in family size. American Economic Review88: 450–477.
2.
AngristJ. D., and KruegerA. B.1991. Does compulsory school attendance affect schooling and earnings?Quarterly Journal of Economics106: 979–1014.
3.
AngristJ. D., and PischkeJ.-S.2009. Mostly Harmless Econometrics: An Empiricist's Companion.Princeton, NJ: Princeton University Press.
4.
BekkerP., KapteynA., and WansbeekT.1987. Consistent sets of estimates for regressions with correlated or uncorrelated measurement errors in arbitrary subsets of all variables. Econometrica55: 1223–1230.
5.
BhalotraS. R., and ClarkeD.2016. The twin instrument. IZA Discussion Paper No. 10405, Institute for the Study of Labor (IZA). http://ftp.iza.org/dp10405.pdf.
6.
BoundJ., JaegerD. A., and BakerR. M.1995. Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak. Journal of the American Statistical Association90: 443–450.
7.
BucklesK. S., and HungermanD. M.2013. Season of birth and later outcomes: Old questions, new answers. Review of Economics and Statistics95: 711–724.
8.
ChernozhukovV., LeeS., and RosenA. M.2013. Intersection bounds: Estimation and inference. Econometrica81: 667–737.
ConleyT. G., HansenC. B., and RossiP. E.2012. Plausibly exogenous. Review of Economics and Statistics94: 260–272.
11.
DahlG. B., and MorettiE.2008. The demand for sons. Review of Economic Studies75: 1085–1120.
12.
HansenL. P.1982. Large sample properties of generalized method of moments estimators. Econometrica50: 1029–1054.
13.
HotzV. J., MullinC. H., and SandersS. G.1997. Bounding causal effects using data from a contaminated natural experiment: Analysing the effects of teenage childbearing. Review of Economic Studies64: 575–603.
14.
KangH., ZhangA., CaiT. T., and SmallD. S.2016. Instrumental variables estimation with some invalid instruments and its application to Mendelian randomization. Journal of the American Statistical Association111: 132–144.
15.
KitagawaT.2015. A test for instrument validity. Econometrica83: 2043–2063.
16.
KlepperS., and LeamerE. E.1984. Consistent sets of estimates for regressions with errors in all variables. Econometrica52: 163–184.
17.
KolesárM., ChettyR., FriedmanJ., GlaeserE., and ImbensG. W.2015. Identification and inference with many invalid instruments. Journal of Business and Economic Statistics33: 474–484.
18.
LeamerE. E.1981. Is it a demand curve, or is it a supply curve? Partial identification through inequality constraints. Review of Economics and Statistics63: 319–327.
19.
ManskiC. F., and PepperJ. V.2000. Monotone instrumental variables: With an application to the returns to schooling. Econometrica68: 997–1010.
20.
ManskiC. F., and PepperJ. V.2009. More on monotone instrumental variables. Econometrics Journal12: S200–S216.
21.
NevoA., and RosenA.2008. Identification with imperfect instruments. Working Paper CWP16/08, Cemmap, Institute for Fiscal Studies. http://www.cemmap.ac.uk/wps/cwp1608.pdf.
22.
NevoA., and RosenA.2012a. Replication data for: Identification with imperfect instruments. Harvard Dataverse.http://hdl.handle.net/1902.1/18721.
23.
NevoA., and RosenA.2012b. Identification with imperfect instruments. Review of Economics and Statistics94: 659–671.
24.
RosenzweigM. R., and WolpinK. I.1980a. Testing the quantity-quality fertility model: The use of twins as a natural experiment. Econometrica48: 227–240.
25.
RosenzweigM. R., and WolpinK. I.1980b. Life-cycle labor supply and fertility: Causal inferences from household models. Journal of Political Economy88: 328–348.
26.
RosenzweigM. R., and WolpinK. I.2000. Natural “natural experiments” in economics. Journal of Economic Literature38: 827–874.
27.
RosenzweigM. R., and ZhangJ.2009. Do population control policies induce more human capital investment? Twins, birth weight and China's “one-child” policy. Review of Economic Studies76: 1149–1174.
RossiP. E., ConleyT. G., and HansenC. B.2012. Replication data for: Plausibly exogenous. Harvard Dataverse.http://hdl.handle.net/1902.1/18022.
30.
RubinD. B.1974. Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology66: 688–701.
31.
SarganJ. D.1958. The estimation of economic relationships using instrumental variables. Econometrica26: 393–415.
32.
SmallD. S.2007. Sensitivity analysis for instrumental variables regression with overidentifying restrictions. Journal of the American Statistical Association102: 1049–1058.
33.
van KippersluisH., and RietveldN.2017. Beyond plausibly exogenous. Tinbergen Institute Discussion Paper TI2017-096/V, Tinbergen Institute. https://papers.tinbergen.nl/17096.pdf.
34.
WisemanN., and SØrensenT. A.2017. Bounds with imperfect instruments: Leveraging the implicit assumption of intransitivity in correlations. IZA Discussion Paper No. 10646, Institute of Labor Economics (IZA). http://ftp.iza.org/dp10646.pdf.
