Aggregated Norton’s equivalent models, with parallel impedance and current injection at different harmonic frequencies are used to model the distribution grid in harmonic studies. These models are derived based on measurements and/or prior knowledge about the grid. The measurement-based distribution (sub-)grid impedance estimation method uses harmonic phasors of 3-phase current and voltage measurements to capture the response of the distribution (sub-)grid before and after an event in the utility side of the grid. However, due to increasing non-linear components in the grid, knowledge about uncertainty in parameters of such equivalent models which intrinsically describe a linear grid becomes important. The aim of this paper is to present two novel methods to calculate the uncertainty of the measurement-based Norton’s equivalent harmonic model of the distribution (sub-)grids as seen from the utility side at the Point of Common Coupling (PCC). The impedance and the uncertainty calculations are demonstrated on a simulated network.