Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cointegration, linear and nonlinear causality

Cointegration, linear and nonlinear causality Purpose – The purpose of this paper is to study the market efficiency, unbiasedness and the direction of causality among four agricultural commodity futures contracts for a forecasting horizon of 28 days, 56 days and 84 days which are traded at National Commodity and Derivatives Exchange Ltd. Design/methodology/approach – To analyse the efficiency of futures market in Indian scenario, we focus on maize, chickpea, soybean and wheat which are among the most important agricultural commodities traded in India. In the first step, Augmented Dickey-Fuller test and nonparametric Phillips-Perron approaches have been used to examine the stationarity of all futures and spot price series. After testing the presence of cointegration in futures and spot series using Johansen’s Cointegration approach, the joint restrictions of β 0 =0, β 1 =1 and β 1 =1 on the cointegrating vectors were imposed to test whether the futures price is an unbiased predictor of spot at contract maturity. In the next step, linear Toda and Yamamoto (1995) and the nonparametric Diks and Panchenko (2006) causality tests were applied to examine the direction of causality. Finally, nonlinear test were applied on the vector error correction model (VECM) residuals to investigate whether any remaining causality is strictly nonlinear in nature. Findings – The results of cointegration tests between futures and spot prices of the selected agricultural commodities indicated a long term relationship do exist in three out of four futures contracts. However, the Wald tests results on the cointegrating vectors indicate markets as inefficient and biased. Further, analysis of short-term relationship using alternate tests of causality do not give consistent results for same commodity series indicating that results may vary due to alternate measures and specifications. Finally, if we consider the results of Diks-Panchenko test on the filtered VECM-residuals, results provide evidence that if cointegration is taken into account; neither spot nor future leads or lags the other consistently. Research limitations/implications – The results are based on the sample of four agricultural futures commodity contracts. The study can be extended to a larger sample of contracts and relative efficiency of each contract can be explored. Originality/value – There are very few studies that have explored the efficiency, unbiasedness and direction of causality using both linear and nonlinear techniques for Indian agriculture commodity futures market for different forecasting horizons. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Agribusiness in Developing and Emerging Economies Emerald Publishing

Cointegration, linear and nonlinear causality

Download PDF
15 pages

Loading next page...
 
/lp/emerald-publishing/cointegration-linear-and-nonlinear-causality-haiheN1UI3
Publisher
Emerald Publishing
Copyright
Copyright © Emerald Group Publishing Limited
ISSN
2044-0839
DOI
10.1108/JADEE-07-2012-0019
Publisher site
See Article on Publisher Site

Abstract

Purpose – The purpose of this paper is to study the market efficiency, unbiasedness and the direction of causality among four agricultural commodity futures contracts for a forecasting horizon of 28 days, 56 days and 84 days which are traded at National Commodity and Derivatives Exchange Ltd. Design/methodology/approach – To analyse the efficiency of futures market in Indian scenario, we focus on maize, chickpea, soybean and wheat which are among the most important agricultural commodities traded in India. In the first step, Augmented Dickey-Fuller test and nonparametric Phillips-Perron approaches have been used to examine the stationarity of all futures and spot price series. After testing the presence of cointegration in futures and spot series using Johansen’s Cointegration approach, the joint restrictions of β 0 =0, β 1 =1 and β 1 =1 on the cointegrating vectors were imposed to test whether the futures price is an unbiased predictor of spot at contract maturity. In the next step, linear Toda and Yamamoto (1995) and the nonparametric Diks and Panchenko (2006) causality tests were applied to examine the direction of causality. Finally, nonlinear test were applied on the vector error correction model (VECM) residuals to investigate whether any remaining causality is strictly nonlinear in nature. Findings – The results of cointegration tests between futures and spot prices of the selected agricultural commodities indicated a long term relationship do exist in three out of four futures contracts. However, the Wald tests results on the cointegrating vectors indicate markets as inefficient and biased. Further, analysis of short-term relationship using alternate tests of causality do not give consistent results for same commodity series indicating that results may vary due to alternate measures and specifications. Finally, if we consider the results of Diks-Panchenko test on the filtered VECM-residuals, results provide evidence that if cointegration is taken into account; neither spot nor future leads or lags the other consistently. Research limitations/implications – The results are based on the sample of four agricultural futures commodity contracts. The study can be extended to a larger sample of contracts and relative efficiency of each contract can be explored. Originality/value – There are very few studies that have explored the efficiency, unbiasedness and direction of causality using both linear and nonlinear techniques for Indian agriculture commodity futures market for different forecasting horizons.

Journal

Journal of Agribusiness in Developing and Emerging Economies – Emerald Publishing

Published: Nov 11, 2014

There are no references for this article.