10 years in my opinion is too short as it might include one property cycle, two property cycles or even none at all. Changing the start year of a 10 year analysis can make the difference between the same suburb being a top performer or a dog.
Longer term is more interesting and you can get rough indications from looking at for sale ads in newspapers 20, 30 or 40 years old. Asking prices are near enough to sale prices in most cases as it's not going to be very exact anyway. Try your local state library for archives.