What is the consequence of the fact that the traditional t-tests are only useful for stationary series and not non stationary series like random walks (2)

The consequence of the fact that traditional t-tests are only useful for stationary series and not non-stationary series like random walks is that they may not accurately capture the true differences between series. Traditional t-tests assume a normal distribution of data that is stable over time, while random walks do not follow this assumption. As a result, traditional t-tests are unable to accurately detect true differences between the means of two series, as the variance between them is not being correctly evaluated. This then leads to inaccurate results, making traditional t-tests unreliable when dealing with non-stationary series like random walks.