The paper “Testing constancy in varying coefficient models” couthored with L. A. Arteaga-Molina and published in the Journal of Econometrics, 222(1), 625-644, 2021 proposes coefficients constancy test in semi-varying coefficient models that only needs to estimate the restricted coefficients under the null hypothesis. The testing procedure resembles in spirit the union-intersection (U-I) parameter stability tests in time series after observations are sorted according to the explanatory variable responsible for the varying coefficients. In recent years, increasing interest has been shown in problems concerning stability of a regression model because changes in economic factors may cause instability of their initial models over a long period of time. For example, technical progress, changes in policies and regulations, or a different economic environment can induce a change among economic variables, even though no change in the parameters of the structural relationship is present. In this context, our test can also be applied to model specification checks of interactive effects in linear regression models. This statistic depends on a trimming parameter that can be chosen by a data-driven calibration method we propose. Because test statistics are not asymptotically pivotal, critical values and p-values are estimated using a bootstrap technique. The bootstrap test is justified under fairly general regularity conditions. Under more restrictive assumptions, the critical values can be tabulated, and trimming is unnecessary. The finite sample properties of the test are investigated by means of Monte Carlo experiments. We also report an application to returns of education modelling.
The paper is available online.