A Monte Carlo Study on the Pitfalls in Determining Deterministic Components in Cointegrating Models

In this paper we examine, by means of Monte Carlo simulation, the properties of the so called 'Pantula principle' for the simultaneous determination of rank and deterministic components in a vector error correction model. Examining the five models contained within the Johansen methodology, we find that the 'Pantula principle' is heavily biased towards choosing model 3 (unrestriced constant) when model 4 (restricted trend) is the true one. We suggest a modification that reduces this bias to an important extent