House price prediction with regression and time series

In this classic example, the tutorial approach is often to treat this as a regression problem to predict house price. However, in the real world we know that house price is time dependant as the house no market rises and falls. How would you implement this aspect into this type of model?

I feel that adding date sold as a feature wouldn’t be very useful as the model wouldn’t see these dates in any new data? But could you add …

datascience prediction price regression time time series