Testing for Conditional Mean Independence with Covariates through Martingale Difference Divergence
This work addresses a crucial problem in statistics for model selection, but it appears incremental as it builds on existing martingale difference divergence methods.
The authors tackled the problem of determining whether additional variables are needed in a regression model by proposing a new multivariate test for conditional mean independence of Y given X conditioning on Z, showing that the estimation error from approximation is negligible under certain assumptions.
As a crucial problem in statistics is to decide whether additional variables are needed in a regression model. We propose a new multivariate test to investigate the conditional mean independence of Y given X conditioning on some known effect Z, i.e., E(Y|X, Z) = E(Y|Z). Assuming that E(Y|Z) and Z are linearly related, we reformulate an equivalent notion of conditional mean independence through transformation, which is approximated in practice. We apply the martingale difference divergence (Shao and Zhang, 2014) to measure conditional mean dependence, and show that the estimation error from approximation is negligible, as it has no impact on the asymptotic distribution of the test statistic under some regularity assumptions. The implementation of our test is demonstrated by both simulations and a financial data example.