Birke, Melanie2007-02-212007-02-212007-02-21http://hdl.handle.net/2003/2329610.17877/DE290R-12785A central limit theorem for the weighted integrated squared error of kernel type estimators of the first two derivatives of a nonparametric regression function is proved by using results for martingale differences and U-statistics. The results focus on the setting of the Nadaraya- Watson estimator but can also be transfered to local polynomial estimates.enCentral limit theoremIntegrated squared errorKernel estimatesLocal polynomial estimateNadaraya-Watson estimateNonparametric regression004report