With models, it's almost always done by trial and error. If a certain model has too little or too much yaw stability, the next version will be similar, but the wing position and/or the dihedral angle will be adjusted very slightly. The math required design an airplane from scratch with guaranteed neutral yaw stability goes way beyond anything that can easily be presented in this forum. There are many factors to consider, such as lateral fuselage area distribution, vertical stabilizer area distribution (rolling moment due to vertical assymetry), wing planform and washout distribution, downwash interaction with the horizontal tail, incidence angles, spiral slipstream effect, and the list goes on.

UNLESS the plane is similar to a Yak 54 or some pattern planes: mid wing, zero thrust angle and zero wing and tail incidence, and a fuselage that is nearly symmetrical vertically about the longitudinal axis. In short, any vertical assymetry can cause undesired (or desired) roll or pitch coupling with yaw.

In general, if it is a low wing model, neutral stability will require some geometric dihedral, and so forth.