Some remarks on the nonlinear Schrödinger equation in the subcritical case