Economists have long been interested in empirically modelling/forecasting the price of staple commodities using both structural and reduced form models. The purpose of this paper is to apply Pesaran’s Autoregressive Distributed Lag/ Error-Correction Bounds model for forecasting the ex-vessel price of lobster on the east coast of Canada using monthly data for the period 1990(1)-2013(12). Since the collapse of the groundfish fishery in the early nineties, crustaceans (particularly lobster) have developed into the most important fisheries in Canada. The advantage of the Pesaran Bounds model is that it is valid regardless of the stationary/probability properties of the variables in question (i.e., ~I(1), ~I(0), or fractionally integrated) and allows the model to capture both short- and long-run price adjustments in the fishery, and importantly, the speed of adjustment to a new equilibrium. Empirically, the base model is an inverse demand equation specified for the ex-vessel price of lobster, lobster landings in Canada and U.S., the U.S. GDP, Canadian/U.S. exchange rate and a Canadian producer price index of finfish. Using four alternative stationarity tests we show a serious lack of consistency in determining the rank of each variable and this supports the Pesaran modelling approach. What is more, using a Boswijk and Urbain test Canadian landings are shown to be endogenous in the model. Econometric procedures are used to address these issues. The interest here is to generate accurate forecasts and predict turning points in ex-vessel prices that will support planning, management and predicting income and welfare effects of Canadian lobster fishermen.