Bootstrap cointegration tests in ARDL models

The paper proposes a new bootstrap approach to Pesaran, Shin, and Smith’s bound tests in conditional equilibrium correction model overcome some typical drawbacks of the latter, such as inconclusive inference distortion size. are worked out under several data-generating processes, including degenerate cases. Monte Carlo simulations confirm better performance relative ones asymptotic F test on independent variables autoregressive distributed lag, or ARDL, model. Empirical applications highlight importance employing appropriate specification provide definitive answers when exploring long-term relationship between economic variables.