Functional coefficient quantile regression model with time-varying loadings

This paper proposes a functional coefficient quantile regression model with heterogeneous and time-varying coefficients factor loadings. Estimation of the is done in two stages. First, we estimate unobserved common factors from linear exogenous covariates. Second, plug-in an affine transformation estimated to obtain model. The parameter estimators are consistent asymptotically normal. application this process cross-section U.S. firms’ excess returns confirms predictive ability firm-specific covariates good performance local estimator coefficients.