I calculated the root mean squared error (RMSE) to compare the simulated values of a hydrologic model with the corresponding observations for three observational datasets.The number of data points in these datasets are 30, 40, and 60.
How to find out if the RMSE values calculated with respect to these datasets are equivalent ? Does the change in the size of the datasets affect the power of the performance metric? How do I detect if it does? Is there a testing mechanism?
Any suggestions ? Thanks.
