Testing Stationarity of a series is an activity which has to be carried out by any quant trader, irrespective of whatever the frequency of data that is being used to develop models. Hence the amount of literature on Unit Roots, is  VERY HUGE.

Here is an interesting and simple way to test it. This note is refreshingly different from the usual stuff that one comes across in Unit Roots. It uses F test as a filter before going to the traditional tests. What does this twist of using F test do ? Well, it increases the power and reduces Type I error of the null hypo that the series is non stationary.

Link : Testing for Unit Roots: What Should Students Be Taught ?

**Abstract :
**Unit-root testing strategies are unnecessarily complicated because they do not exploit prior knowledge of the growth status of the time series, they worry about unrealistic outcomes, and they double- or triple-test for unit roots. The authors provide a testing strategy that cuts through these complications and so facilitates teaching this dimension of the unit-root phenomenon. F tests are used as a vehicle for understanding, but t tests are recommended in the end, consistent with common practice.