Concerning the use of the Kramers-Kronig transforms for the validation of impedance data