|dc.description.abstract||This dissertation analyzes the Dungeness crab prices and quantities, which is conducted within three essays. The first essay studies the relationships among the West Coast Dungeness crab landing prices and quantities using cointegration analysis and directed acyclic graphs. The forecast tests are added to determine the number of cointegrating rank. Directed acyclic graphs are estimated using different algorithms for comparison and are used to discover the causality of the crab markets. The four states? crab prices are strongly connected contemporaneously. The price-quantity relationships exist among the California, Oregon and Washington markets because of their tri-state Dungeness crab project. The Alaska quantity does not affect and is not affected by the other prices and quantities possibly due to stock collapse in some areas of Alaska.
The second essay uses the three models to explore the prequential relationships among the West Coast states? Dungeness crab fisheries. A random walk and the 1-lag VAR outperform the 2-lag VAR. Most series in the random walk and the l-lag VAR are well-calibrated. For the Dungeness crab quantities, the random walk does slightly better than the 1-lag VAR; the 1-lag VAR dominates the random walk for the crab prices. The results are consistent with the literature on the Dungeness crab movement patterns. Information about the crab fishery management decision making are provided in this essay.
The third essay estimates the Dungeness crab yield insurance premiums and the probabilities of the indemnities being paid to the crab fishermen in each western coastal state using cointegration analysis, goodness-of-fit tests, and Monte Carlo simulation. The lognormal distribution provides the best-fit for the Alaska crab yield and the logistic for the Oregon, Washington, and California yields, respectively. The log-logistic is found to be the best-fit for each state?s prices. At 50%, 60%, 70%, and 80% yield coverage levels, Alaska has the highest insurance premiums and the highest probability of paying the indemnities, followed by California, and then Washington or Oregon.||