If you are unsure, you can run Monte Carlo simulations to evaluate how your model performs in a controlled environment given settings that are relevant in your empirical application. The answer will likely depend on the number of parameters in your model, and the actual parameter values in the DGP.