The RATS Software Forum

Dear Tom,

How do I get the critical values when performing the Lee and Strazicich (2003) LM unit root tests with multiple breaks? In their papers the critical values are given for only a handful of cases for one and two breaks. Even for these cases my lambdas are different from those available in their papers, where my second break occurs near the end of the data.

Regards
Hashem

The asymptotics for the LS test and most similar procedures are based on the break(s) being somewhere in the middle of the data set, and a completely different theory would apply near the ends.