Uniform inference in linear error-in-variables models: Divide-and-conquer

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 nonzero. 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 <i>q</i> and cash flow, we find time periods in which the effect of Tobin’s <i>q</i> is not statistically different from zero. The implausibly large higher-order moment estimates in these periods disappear when using the proposed estimator.