Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

A Study of the Uncertainty of Farm-product Prices by Entropy Model using Probability Distribution for Monthly Prices

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  Paper number  131620093,  2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: @2013
Authors:   Sangkyu Eun, Jeong-Jae Lee, Yeongjoung Bae
Keywords:   prediction of agricultural prices fluctuation probability distribution entropy model

Abstract. Commodity prices are affected by supply and demand conditions and other uncertain factors which are difficult to predict. Particularly, there is greater variability in farm-product prices as they are affected by long-term unpredictable factors such as weather conditions, demands and so on, because it takes long periods of time to harvest crops. To predict price variations, various models such as ‘Price Fluctuation Probability based on Reliability Analysis Method’ (PFPRA), ‘Prediction of Price Change using Signal Sampling Method’(PPCSS), ‘Establishment of Range of Risk in accordance with Price Change (ERRPC)’ have been developed.

The PFPRA model has a problem of lower degree of accuracy in the prediction of price fluctuation for medium and long term periods compared to that of short term periods. The PPCSS model is good for the prediction of price fluctuations for long term periods; however, statistical verification on its results is difficult to know. The ERRPC model has problems with different elements that affect the pricing decision, which makes it difficult to use the model for general prediction purposes.

From the theory of statistical thermodynamics and information, it is possible to quantify the complexity systems for their phenomena when ‘Entropy Model (EM)’ is used. The EM is used as a tool to describe complex social and economic systems, and analyze them. Particularly, the EM has the merit of measuring the uncertainty of a system as a whole using observed partial data of the target system.

For this research, information was collected on the price changes for agricultural products, which was then used to quantify the uncertainty of factors contributing to the pricing decision. Finally, a model, which presents the uncertainty in terms of entropy values, is proposed. This model improves the problems which existing models have on the prediction of farm-product prices, and can be applied to a prediction model for general price change.

(Download PDF)    (Export to EndNotes)