Modeling in Natural Resource Economics: Forecasting Electricity Demand, and Simulatinglandowner Response to Wildfire Risk

This work presents three modeling techniques appliedto natural resource economics. Chapter I makes use ofordinary least-squares regression analysis (OLS) toinvestigate the effects of climate warming onelectricity demand. Using hourly models and data forthe U. S. mid-Atlantic, the model predicts a 4.6%Chapter II offers an introduction to modeling withartificial neural networks (ANNs) for researchersfamiliar with OLS techniques. The general form of anANN is shown to be equivalent to a nested series ofOLS models. The forecasting exercise from Chapter Iis recast in the ANN framework to provide an exampleof applying ANNs to economic studies, including ananalysis of the forecast residuals. Chapter III presents a model of forest-landownerbehavior in the context of managing timberlandsubject to wildfire. The model is used to examine theimpacts of externalities associated with fire risk:(1) the public-good aspect of privately funded fireprevention, and (2) moral hazard associated withpublicly funded fire suppression. Chapter IIIconcludes with a consideration of policy options foraddressing these externalities.