Uniform Inference in Linear Error-in-Variables Models: Divide-and-Conquer
arXiv:2301.04439
Abstract
It is customary to estimate error-in-variables models using higher-order moments of observables. This moments-based estimator is consistent only when the coefficient of the latent regressor is assumed to be non-zero. We develop a new estimator based on the divide-and-conquer principle that is consistent for any value of the coefficient of the latent regressor. In an application on the relation between investment, (mismeasured) Tobin's and cash flow, we find time periods in which the effect of Tobin's is not statistically different from zero. The implausibly large higher-order moment estimates in these periods disappear when using the proposed estimator.
32 pages, 4 figures, 5 tables