WSEAS Transactions on Business and Economics
Print ISSN: 1109-9526, E-ISSN: 2224-2899
Volume 13, 2016
Efficiency and Predictability on International Soybean Oil Prices
Authors: , , , ,
Abstract: In analysing the efficiency, non-linearity and predictability of soybean [Glycine max (L.) Merr.] oil price series structural breaks were identified. The efficiency of the market in its weak form was then analysed using automatic variance ratio for small samples. Group Method of Data Handling (GMDH) polynomial neural networks, the Box Jenkins Method and Genetic Algorithms were tested for their ability to predict return on a monthly basis. The analysis suggested an inefficiency in weak form and some predictability of the market. The Diebold-Mariano, used to discriminate amongst models according to their accuracy, indicated that the combined use of linear (ARIMA) and non-linear (e.g., multilayered and self-organizing artificial neural networks of the GMDH and Genetic Algorithm type) techniques significantly improved the market prediction.
Search Articles
Pages: 444-456
WSEAS Transactions on Business and Economics, ISSN / E-ISSN: 1109-9526 / 2224-2899, Volume 13, 2016, Art. #40