Generalized kernel regularized least squares estimator with parametric error covariance

Abstract A two-step estimator of a nonparametric regression function via Kernel regularized least squares (KRLS) with parametric error covariance is proposed. The KRLS, not considering any information in the covariance, improved by incorporating allowing for both heteroskedasticity and autocorrelation, estimating function. two step procedure used, where first step, estimated using KRLS residuals second transformed model KRLS. Theoretical results including bias, variance, asymptotics are derived. Simulation show that proposed outperforms heteroskedastic errors autocorrelated cases. An empirical example illustrated an airline cost under random effects correlated errors. derivatives evaluated, average partial inputs determined application.