A0233
Title: Nonparametric significance testing in measurement error models
Authors: Luke Taylor - London School of Economics (United Kingdom) [presenting]
Abstract: A nonparametric significance test for regression models with measurement error in the regressors is developed. To the best of our knowledge, this is the first test of its kind. We use a `semi-smoothing' approach with nonparametric deconvolution estimators and show that our test is able to overcome the slow rates of convergence associated with such estimators. In particular, our test is able to detect local alternatives at the $\sqrt{n}$-rate. We derive the asymptotic distribution under weakly dependent data and provide a bootstrap procedure. We also highlight the finite sample performance of the test through a Monte Carlo study. Finally, we discuss two empirical applications. The first considers the effect of cognitive ability on a range of socio-economic variables: income, life satisfaction, health and risk aversion. The second uses time series data to investigate whether future inflation expectations are able to stimulate current consumption; an important policy question when nominal interest rates approach the zero lower bound.
