This article presents a new Stata command, miivfind, that implements an algorithm developed by Bollen and Bauer (2004, Sociological Methods and Research 32: 425–452) to find the model-implied instrumental variables (MIIVs) from an identified structural equation model. MIIVs allow researchers to draw on instrumental-variable estimators, such as two-stage least-squares estimators, to obtain estimates for the parameters of a hypothesized structural equation model. It can be difficult to identify MIIVs by inspection of either a diagram of the model or the model equations. Two examples are provided that illustrate the use of miivfind to identify MIIVs and some of the advantages of a MIIV estimator as compared with a maximum likelihood estimator. By assisting in the process of finding MIIVs, miivfind facilitates the use of an alternative class of estimators, instrumental-variable estimators, to the standard maximum-likelihood and asymptotic-distribution free estimators available for structural equation models.
AcemogluD., JohnsonS., and RobinsonJ. A.2001. The colonial origins of comparative development: An empirical investigation. American Economic Review91: 1369–1401.
2.
AngristJ. D., and PischkeJ.-S.2009. Mostly Harmless Econometrics: An Empiricist's Companion.Princeton, NJ: Princeton University Press.
3.
AnselinL.1988. Spatial Econometrics: Methods and Models.Dordrecht: Kluwer Academic Publishers.
4.
ArellanoM., and BondS.1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies58: 277–297.
5.
BasmannR. L.1960. On finite sample distributions of generalized classical linear identifiability test statistics. Journal of the American Statistical Association55: 650–659.
6.
BollenK. A.1987. Total, direct, and indirect effects in structural equation models. Sociological Methodology17: 37–69.
7.
BollenK. A.1989. Structural Equations with Latent Variables.New York: Wiley.
8.
BollenK. A.1996a. A limited-information estimator for LISREL models with and without heteroscedastic errors. In Advanced Structural Equation Modeling Techniques, ed. MarcoulidesG. A., and SchumackerR. E., 227–241. Mahwah, NJ: Lawrence Erlbaum.
9.
BollenK. A.1996b. An alternative two stage least squares (2SLS) estimator for latent variable equations. Psychometrika61: 109–121.
10.
BollenK. A.2001. Two-stage least squares and latent variable models: Simultaneous estimation and robustness to misspecifications. In Structural Equation Modeling: Present and Future: A Festschrift in Honor of Karl Jöreskog, ed. CudeckR., ToitS. D., and SörbomD., 119–138. Lincolnwood, IL: Scientific Software.
11.
BollenK. A., and BauerD. J.2004. Automating the selection of model-implied instrumental variables. Sociological Methods and Research32: 425–452.
12.
BollenK. A., and BiesanzJ. C.2002. A note on a two-stage least squares estimator for higher-order factor analyses. Sociological Methods and Research30: 568–579.
13.
BollenK. A., KirbyJ. B., CurranP. J., PaxtonP. M., and ChenF.2007. Latent variable models under misspecification: Two-stage least squares (2SLS) and maximum likelihood (ML) estimators. Sociological Methods and Research36: 48–86.
14.
BollenK. A., and Maydeu-OlivaresA.2007. A polychoric instrumental variable (PIV) estimator for structural equation models with categorical variables. Psychometrika72: 309–326.
15.
CameronA. C., and TrivediP. K.2005. Microeconometrics: Methods and Applications.New York: Cambridge University Press.
16.
CameronA. C., and TrivediP. K.2010. Microeconometrics Using Stata.Rev. ed. College Station, TX: Stata Press.
17.
HägglundG.1982. Factor analysis by instrumental variables methods. Psychometrika47: 209–222.
18.
HeckmanJ. J., UrzuaS., and VytlacilE. J.2006. Understanding instrumental variables in models with essential heterogeneity. Review of Economics and Statistics88: 389–432.
19.
ImbensG. W., and AngristJ. D.1994. Identification and estimation of local average treatment effects. Econometrica62: 467–475.
20.
JöreskogK. G.1977. Structural equation models in the social sciences: Specification, estimation, and testing. InApplications of Statistics, ed. KrishnaiahP. R., 265–287. Amsterdam: North-Holland.
21.
JöreskogK. G.1983. Factor analysis as an error-in-variables models. In Principles of Modern Psychological Measurement: A Festschrift for Frederic M. Lord, ed. WainerH., and MessickS., 185–196. Hillsdale, NJ: Lawrence Erlbaum.
22.
JöreskogK. G., and SörbomD.1993. LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language.Lincolnwood, IL: Scientific Software International.
23.
KirbyJ. B., and BollenK. A.2009. Using instrumental variable (IV) tests to evaluate model specification in latent variable structural equation models. Sociological Methodology39: 327–355.
24.
MadanskyA.1964. Instrumental variables in factor analysis. Psychometrika29: 105–113.
25.
SarganJ. D.1958. The estimation of economic relationships using instrumental variables. Econometrica26: 393–415.
26.
StockJ. H., and YogoM.2005. Testing for weak instruments in linear IV regression. In Identification and Inference for Econometric Models: Essays in Honor of Thomas Rothenberg, ed. AndrewsD. W. K., and StockJ. H., 80–108. New York: Cambridge University Press.
27.
WooldridgeJ. M.2010. Econometric Analysis of Cross Section and Panel Data. 2nd ed. Cambridge, MA: MIT Press.
