An analysis of commodity price dynamics with focus on the price of salmon
MetadataShow full item record
- PhD theses (TN-IØRP) 
Original versionAn analysis of commodity price dynamics with focus on the price of salmon by Atle Øglend, Stavanger : University of Stavanger, Faculty of Science and Technology, Department of Industrial Economics, Risk Management and Planning 2010 (PhD thesis UiS, no. 99)
This thesis is concerned with studying the short run dynamics of commodity prices. The industry of interest and primary study is Norwegian Aquaculture, with the price of farmed salmon as the main data-set. Even though most of the cases studied are related to the salmon market, it is my hope that some of the insights and results can be applied to a more general set of agricultural or conventional commodities. This thesis falls in line with a large collection of research papers and thesis’ on the Norwegian aquaculture industry. Motivated by dissecting what has largely been a highly successful growth industry, coupled with availability of detailed high quality data, a great deal of economic research on the industry has been conducted. Much of this research is related to long run supply side effects. A large low frequency panel data set has laid the ground for successful economic research into amongst other productivity effects in the industry. Due to a lack of high frequency data, the short run effects have been less studied. The only reliable high frequency data available is price data. This thesis contributes to the body of research on the industry by focusing on the short run price dynamics of the commodity. In addition to studying short run effects, the thesis introduces tools originally used in finance to study the price data. Incorporating both traditional economic analysis and finance is relevant when doing short run price analysis, and provides an alternative angle for looking at the commodity market. Due to the lack of detailed high frequency data on state variables other than price, the thesis applies non-structural time series analysis as the method for empirical analysis. This necessarily restricts direct inference of causality relationships. However, nonstructural time series analysis provides a large battery of models to reliably and thoroughly describe the dynamics of the series studied. The detailed output from time series models are used in combination with knowledge of predictable relative changes in underlying state variables to understand the market dynamics.