Why is it, in general, important to test for non-stationarity in time series data before attempting to build an empirical model?
a) Stationary data simplifies model building
b) Non-stationary data is more predictable
c) Non-stationary data violates model assumptions
d) Stationary data is always more accurate