Working Paper: NBER ID: w13134
Authors: Kenneth D. West; Kafu Wong; Stanislav Anatolyev
Abstract: We propose and evaluate a technique for instrumental variables estimation of linear models with conditional heteroskedasticity. The technique uses approximating parametric models for the projection of right hand side variables onto the instrument space, and for conditional heteroskedasticity and serial correlation of the disturbance. Use of parametric models allows one to exploit information in all lags of instruments, unconstrained by degrees of freedom limitations. Analytical calculations and simulations indicate that there sometimes are large asymptotic and finite sample efficiency gains relative to conventional estimators (Hansen (1982)), and modest gains or losses depending on data generating process and sample size relative to quasi-maximum likelihood. These results are robust to minor misspecification of the parametric models used by our estimator.
Keywords: Instrumental Variables; Heteroskedasticity; GMM; Econometrics
JEL Codes: C13; C32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| proposed estimator for instrumental variables (C26) | large asymptotic efficiency gains (D61) |
| large asymptotic efficiency gains (D61) | performance in hypothesis testing (C12) |
| large asymptotic efficiency gains (D61) | mean squared error (C20) |
| proposed estimator (C51) | performance compared to maximum likelihood estimators (C51) |
| efficiency of new estimator (C51) | efficiency of conventional GMM (C51) |
| use of additional instruments (C36) | increase asymptotic efficiency (C51) |
| conditional heteroskedastic disturbances (C21) | efficiency of estimators (C51) |