I i have applied NG Perron's modified tests to my timer series data. But when I tried to interpret test statistics I saw that, the decision given by MZa and MAzt contradicts with MSB and MPT.
Is there an assumption which I should know, that would cause this diversion
Thanks..