Ten lectures on wavelets
The nature of statistical learning theory
The nature of statistical learning theory
| Hi-index | 0.00 |
Accurate forecasts of cash market prices for agricultural commodities are of a great value to market participants who are looking to reduce the risk involved in buying and selling commodities. This paper discusses the use of "The Synthesis" model to forecast cash market prices for corn or any other agricultural commodities four months in advance. The Synthesis does not only forecast a single price realization four months ahead but also forecasts the whole signal up to four months in the future. Predicting a continuous signal at a weekly temporal resolution provides the best forecasts of the resolutions considered. The Synthesis is a synthetic model of wavelets and Support Vector Machines (SVMs). The wavelet decomposition is used to break down the input and the output signals into different temporal resolution then for each consistent set of input and output signal, an SVM machine is employed to unfold the pattern at this particular resolution. The Synthesis has shown good results in comparison to previous and traditional methods.