Dynamic Calibration of Price-Trend Parameters for Commodity Futures Trading

• Type: Article in International Conference Proceedings Book
• Date: 1992

Authors

José Sarsfield Cabral (Author) FEUP; Rui Guimarães (Author) FEUP

Conference proceedings International

Title: Comunicação apresentada na EURO XII / TIMS Conference

EURO Xll / TIMS Conference

Helsinquia, Finlândia

Other information

Resumo (PT): The paper addresses the problem of buying agricultural commodities through the futures markets and deals specifically with a heuristic rule developed for the scenario described as ¿purchasing under a deadline'. The rule is based on short--term forecasts produced by Taylor¿s price-trend model. Previous work applied to CBOT corn futures market showed that the autocorrelation parameters af the process generating the daily returns were nearly stable over bme and hence could be estimated using a static procedure. However, a more recent study conduced on the CBOT soybean futures market revealed a clear instability of those parameters, which impaired a successful application of the purchasing rule. In this paper the authors propose a procedure for calibrating dynamically the price-trend model parameters. This procedure involves the continuous CUSUM monitoring of the purchasing results and derives the parameter estimates from exponentially smoothed sample autocorrelation coefficients of the daily returns. The procedure was tested using the 1972-87 CBOT daily soybean futures closing prices. The results suggest that it leads to a significant improvement on the purchasing results derived from the previously adopted static parameter calibra-tion procedure.

Abstract (EN): The paper addresses the problem of buying agricultural commodities through the futures markets and deals specifically with a heuristic rule developed for the scenario described as ¿purchasing under a deadline'. The rule is based on short--term forecasts produced by Taylor¿s price-trend model. Previous work applied to CBOT corn futures market showed that the autocorrelation parameters af the process generating the daily returns were nearly stable over bme and hence could be estimated using a static procedure. However, a more recent study conduced on the CBOT soybean futures market revealed a clear instability of those parameters, which impaired a successful application of the purchasing rule. In this paper the authors propose a procedure for calibrating dynamically the price-trend model parameters. This procedure involves the continuous CUSUM monitoring of the purchasing results and derives the parameter estimates from exponentially smoothed sample autocorrelation coefficients of the daily returns. The procedure was tested using the 1972-87 CBOT daily soybean futures closing prices. The results suggest that it leads to a significant improvement on the purchasing results derived from the previously adopted static parameter calibra-tion procedure.

Language: Portuguese

Type (Professor's evaluation): Scientific