Title :
Commodity trading with a two year old
Author_Institution :
Martingale Res. Corp., Allen, TX, USA
Abstract :
Summary form only given. Continuing work to train an artificial neural network (ANN) to recognize a long/short (buy/sell) pattern for the live cattle commodity futures market was discussed. The back-propagation of errors algorithm was used to encode the relationship between the long/sell desired output and fundamental indicators plus technical variables into the ANN. Trained on two years of past data, the ANN was able to predict long/short market positions for 5 months in the future that would have made $2752 profit on an investment of $800
Keywords :
commodity trading; investment; learning systems; neural nets; ANN; artificial neural network; back-propagation; errors algorithm; investment; live cattle commodity futures market; long/sell desired output; long/short market positions; profit; Artificial neural networks; Cows; Handwriting recognition; Investments; Marine vehicles; Multi-layer neural network; Nearest neighbor searches; Neural networks; Pattern recognition; Radial basis function networks;
Conference_Titel :
Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-0164-1
DOI :
10.1109/IJCNN.1991.155498
