It's kind of analogous to ionic bonding in chemistry. Take Sodium (Na) and Chlorine (Cl), for example.
Generally, students acquire the misconception that when Na and Cl react to form NaCl (commmon table salt), Cl simply removes an electron from Na, forming Na+ and Cl-.
But that's not strictly true.
Electrons actually exist in probabilistic regions, meaning there's a certain chance that an electron is in any position, rather than a defined, singular location. Because of the Octet rule (and/or other reasons), the chlorine atom - having 7 valence electrons - much more often takes on an 8th electron rather than losing all seven to go back to 0 valence electrons (which would then mean that it actually had 8 again, but you get the idea). Applying this to NaCl, the individual chlorine molecule attracts the electron so much more strongly than Na that the electron is extremely likely to be "part of" the chlorine atom at any moment. This inequality is so pronounced that we observe the Cl ion to have a negative charge. This post has some more in-depth information, as does the rest of its host question.
To bring this rant back to Music Theory, no, there are no hard-and-dry rules on how the 7th degree, or any scale degree, resolves. But we observe that nearly every time, the 7th scale degree resolves to the octave rather than all the way back down to the lower octave. A bunch of explanations exist for this, and as others have noted, smaller movements tend to be observed as "smoother" resolutions, but my point is that the 7th degree is commonly said to resolve up to the octave because that's how it happens nearly every time. If you feel it works better down to the octave, by all means give it a try, but recognise that you may be mistaking the resolution's effect with that of the effect of having the lower bass note, and that you may achieve effects that you did not intend or did not desire.