Proper motion data, even the really good stuff we've been getting from the astrometric satellites lately, isn't accurate enough to be able to run the movie backwards for more than a few hundred thousand years. The quality of the data also degrades rapidly the further you get from earth; no more than a couple of hundred ly, we can only sample the dynamics of a tiny portion of the galaxy and only the brightest stars at the limit of that range. At greater distances we get Doppler radial velocities, visual and radio, but no proper motions.

Add to this that stellar orbits about the galactic COM are not Keplerian ellipses, but rosettes, (that dark matter makes galactic rotation appear non-Newtonian). Orbits are highly perturbed by passing stars and massive clouds so the system is highly chaotic. Even the sun porpoises in and out the other side of the galactic plane in an oscillating sinusoid. Tracing back 20-odd orbits is out of the question.

By the way, it's (very approximately) 250 Myr/orbit, divided into 4.5 Gyr. The original cluster which the sun appeared in probably stayed together for only a few orbits, shedding stars as it went along, until it entirely evaporated. It is not impossible the sun may be a member of an association (a group of separated stars of similar proper motion traveling together, and hence, of probable joint origin) but it is not likely in my opinion.

Like the earth's geology, at long eough time scales, the galaxy is as turbulent and chaotic as a boiling pot of stew.