Use this URL to cite or link to this record in EThOS:
Title: Forecasting of daily dynamic hedge ratio in agricultural and commodities' futures markets : evidence from Garch models
Author: Zhang, Yuanyuan
ISNI:       0000 0004 2726 7522
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis investigates the predictive power of six bivariate GARCH-CCC (constant conditional correlation) models; the GARCH (1, 1), BEKK GARCH (1, 1), GARCH-X (1, 1), BEKK-X (1, 1), GARCH-GJR (1, 1) and QGARCH (1, 1) based on both normal and student’s t distributions. Empirical investigations are conducted by forecasting the daily hedge ratios from agricultural futures markets using one-step-ahead over 1 year and 2 year out-of-sample period. The forecasting of OHR in agricultural and commodities’ futures markets has not been studied thoroughly and few publications are available in literature. My work enriches the literature and will hopefully provide guidance for hedging in these markets. To forecast the OHR, we apply data from three storable commodities, coffee, wheat and soybean and two non-storable commodities, live cattle and live hog. Four tests are conducted to evaluate the forecasting errors of out-of-sample forecasted return of the portfolio based on the forecasted OHR. Our study shows that the asymmetric GARCH model outperforms other models, and the standard GARCH is the weakest for 1-year forecast. However, the standard GARCH model performs well for 2-year forecast of live cattle with student’s t distributed residuals. More generally, the BEKK and asymmetric GJR and QGARCH models are recommended to forecast OHR on both 1-year and 2-year horizons with normal and student’s t distributions for storable products and the asymmetric models for non-storable commodities. Furthermore, our study demonstrates that the predictive power of GARCH models depends on the distribution of residuals, the commodity and also the length of the forecast horizons. This result is consistent with the those from Poon and Granger (2003) and Chen (2003). Given accurately forecasted OHR, investors can determine appropriate hedging strategies for portfolio management to reduce or transfer risks, and prepare for the capital needed for hedging.
Supervisor: Choudhry, Taufiq Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HD28 Management. Industrial Management