Testing for error correlation in partially functional linear regression models

This paper investigates the series correlation test for the partially functional linear regression model (PFLRM). PFLRM relates a scalar response to the predictor variable that is composed of multivariate vectors and random functions. We propose two test statistics for series correlation in PFLRM and derive their asymptotic distributions under the null hypothesis. Numerical studies with finite sample size reveal that the proposed statistics have good performances. Finally, we explore the application of the tests on the electricity consumption data which shows that our test statistics are effective and stable.