My belief is that the values for running torque (assuming Drawoh has guessed right) are determined by test with controlled conditions.
I don't remember seeing a way to calculate it, other than extrapolate the test data for a thread size not in the table but all other conditions being met (material, finish, lubricants, etc).
But usually a thread torque table is very thorough and doesn't leave out any of the common sizes.
Can you tell us what thread form you are using? It is possible that a running torque table already exists, but we can't find it unless you tell us what it is.
STF