Empirical Modelling of Canadian Petroleum Exploration Activity
This paper provides an empirical model of Canadian petroleum exploration activity. It represents an improvement over existing models in the sense that it satisfies a number of design criteria which the latter fail to satisfy. The replication and evaluation of existing models which have not been rigorously tested in their published or reported form is treated as an important part of the search for an adequate specification. When existing models are subjected to diagnostic tests, their regression residuals fail to have the desirable white noise properties and the model parameters show evidence of instability. Thus, evidence for misspecification is produced which renders invalid much of the inference made on the basis of these models.
The existing theory of exploration activity has led to the use of constructed data for unobservable variables and to highly restrictive equation specifications which do not provide adequate representations of the data. The use of constructed data imposes severe restrictions on the determinants of exploration behaviour, and particularly on the way in which firms form expectations of future returns from exploration. Consequently, we provide alternative theoretical guidelines which place fewer restrictions on the data and allow for the possibility that firms face credit constraints and have more limited planning horizons than is implied in the existing literature.
2. Alberta Exploration Activity Data and Canadian Petroleum Policy
2.1 The Data and Data Sources
2.2 Canadian Petroleum Policy 1947-82
Appendix to Chapter 2: Sample Reserve Price Calculations
3. The Empirical Literature
3.1 Variable Profit-Function Models
3.2 Problems with the Use of Constructed Price Data
3.3 Testing of Scarfe and Rilkoff (1984)
3.4 Concluding Remarks
4. An Investment Approach to Exploration Expenditures
4.1 An Investment Framework for Modelling
4.2 An Empirical Model of Exploration Expenditures
4.3 Concluding Remarks