A note on Measurement Error and Euler Equations: an Alternative to Log-Linear Approximations

Empirical research based on panel data has to pay special attention to measurement errors. Utility maximization often yields nonlinear decision rules in which measurement errors enter in a multiplicative way. The usual strategy to deal with them consists of taking log-linear approximations of the equations to estímate. The expression to be estimated then includes a new error component and the estimators could be biased and inconsistent. We describe one particular parameterization that avoids linearizing the equation we want to estimate.