Finite-Sample Inference Methods for Simultaneous Equations and Models with Unobserved and Generated Regressors

We propose finite sample tests and confidence sets for models with unobserved and generated regressors as well as various models estimated by instrumental variables methods. The validity of the procedures is unaffected by the presence of identification problems or \"weak instruments\", so no detection of such problems is required. We study two distinct approaches for various models considered by Pagan (1984). The first one is an instrument substitution method which generalizes an approach proposed by Anderson and Rubin (1949) and Fuller (1987) for different (although related) problems, while the second one is based on splitting the sample. The instrument substitution method uses the instruments directly, instead of generated regressors, in order to test hypotheses about the \"structural parameters\" of interest and build confidence sets. The second approach relies on \"generated regressors\", which allows a gain in degrees of freedom, and a sample split technique. For inference about general possibly nonlinear transformations of model parameters, projection techniques are proposed. A distributional theory is obtained under the assumptions of Gaussian errors and strictly exogenous regressors. We show that the various tests and confidence sets proposed are (locally) \"asymptotically valid\" under much weaker assumptions. The properties of the tests proposed are examined in simulation experiments. In general, they outperform the usual asymptotic inference methods in terms of both reliability and power. Finally, the techniques suggested are applied to a model of Tobin’s q and to a model of academic performance.

Predictors and Consequences of Simultaneous Alcohol and Cannabis Use in Adolescents

Background/Aims: The simultaneous use of alcohol and cannabis is common among adolescents, but has been little studied. In this study, we examine predictors and consequences of this behavior in a population-based sample of high school students. Method: Self-reports were obtained from students in Quebec (Canada) followed throughout high school (N=6589). Logistic regressions were used to test the association between individual, family, and peerrelated predictors in grades 7–8 and simultaneous alcohol and cannabis use in grade 10, as well as between simultaneous alcohol and cannabis use in grade 10 and experiencing 3 or more substance-related problems of various types (legal, physical, etc.) in grade 11. Results: Most predictors in grades 7–8 were associated with simultaneous alcohol and cannabis use in grade 10. Only variables reflecting early-onset substance use involvement — alcohol intoxication, cannabis use, and drug use by close friend(s) — remained predictive in a multivariate model. Simultaneous alcohol and cannabis use was associated with increased substance-related problems in grade 11, above and beyond baseline problems and the concurrent use of the two substances in separate episodes in grade 10. Conclusions: Simultaneous alcohol and cannabis use 1) is anticipated by multiple psychosocial risk factors which come together with individual and peer substance use in early high school and 2) is independently predictive of subsequent substance-related problems. Providing adolescents with adequate information regarding the potential harm of simultaneous use may be a useful prevention strategy.