We propose a new testing strategy to test the unit root hypothesis when the linear deterministic time trend is uncertain. Our strategy is obtained by combining the union-based rejection rule and the M-estimation method considering Johnson SU as a reference density. Simulation results demonstrate that our method exhibits good finite sample performance in terms of power and maintains the test size correctly regardless of the time trend’s presence. The proposed method outperforms other strategies in the case of heavy-tailed error processes. We employ the strategy on India’s nominal monthly interest rate and observe superior performance compared to existing methods.
No takes yet. Share an insight, caveat, or question.
Kar et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: