Why you cannot take non-stationary data to solve the time series Problem?

o Most models assume stationary data. In other words, standard techniques are invalid if data is "NON-STATIONARY".

o Autocorrelation may result due to "NON-STATIONARY".

o Non-stationary processes are a random walk with or without a drift (a slow, steady change).

o Deterministic trends (trends that are constant, positive or negative, independent of time for the whole life of the series).