Analysis of Herd Behavior In Commodity Futures Markets

The herd behavior which can be defined as ignoring self-opinions and imitating the other investors in investment decision making process can be a fundamental characteristic of financial markets in crisis periods. The main purpose of the study is to examine the possible existence of the herd behavior in the commodity futures markets for the period of 06.01.1998-07.06.2018. In this study, the methodologies of Christie and Huang (1995) and Chang, Cheng, and Khorana (2000), which are based on the cross-sectional variability of commodity futures contracts, and the Hwang and salmon (2004) method which is based on the cross-sectional variability of beta coefficients were implemented. As a result, the evidence obtained from the method of Christie and Huang (1995), which assumed a linear relation between cross-sectional variability and market returns, does not show herd behavior. On the other hand, the evidence obtained from the method of Chang, Cheng and Khorana (2000), who argue a nonlinear relation supports the hypothesis that there is a herd behavior in the commodity futures market and shows that the relationship between cross sectional variability and high and low rates of return is nonlinear. In particular, it is seen that the effects of herd behavior are more pronounced in the bullish periods of in the market. Furthermore, the findings obtained from the methodology based on the variability of the betas also show the presence of the herd behavior in certain periods.