A test of the Samuelson Hypothesis using realized range

Abstract

This study examines the Samuelson Hypothesis, which postulates that futures price volatility increases as the futures contract approaches its expiration. Investigating intraday data and drawing on the recently developed concept of realized range, this study provides empirical evidence regarding the Samuelson Hypothesis for 14 agricultural, metal, energy, and financial futures markets in six futures exchanges. While utilizing a nonparametric test, a simple linear regression model and a system of seemingly unrelated regressions, the study finds strong support for the Samuelson Hypothesis in agricultural futures. In contrast, no support for the Samuelson Hypothesis is observed in any of the metal and financial futures. © 2008 Wiley Periodicals, Inc. Jrl Fut Mark 28:680–696, 2008

Asymmetric impacts of fundamentals on the natural gas futures volatility: An augmented GARCH approach

“…See, for example,Milonas (1986),Chatrath et al (2002),Duong and Kalev (2008),Kalev and Duong (2008),Karali et al (2010a),Karali et al (2010b) andKarali and Thurman (2010).…”

Maturity effects in the Mexican interest rate futures market

“…More recently, Duong and Kalev (2008) and Kalev and Duong (2008), using intraday data from different futures markets, found strong support for the Samuelson hypothesis in agricultural futures. However, they concluded that the hypothesis does not hold for other futures contracts.…”

“…The gain in efficiency of the SUR estimators over single equation OLS would be large when contemporaneous disturbance terms in different equations are highly correlated and when the independent variables in different equations are not highly correlated (Zellner & Huang, 1962). The arrangement of the data in Kalev and Duong (2008) perfectly fits with the SUR method as time to maturity is different for each time series and disturbances are expected to be contemporaneously correlated. However, when the regression equations are such that the explanatory variables are the same for all the equations, as it is in our case, then GLS is equivalent to equation-by-equation OLS.…”

“…However, the arrangement by days to maturity is necessary for a panel data analysis. 4 To test the presence of maturity effects in a set of contracts with the same underlying, Kalev and Duong (2008) use Seemingly Unrelated Regression (SUR) estimation. The gain in efficiency of the SUR estimators over single equation OLS would be large when contemporaneous disturbance terms in different equations are highly correlated and when the independent variables in different equations are not highly correlated (Zellner & Huang, 1962).…”

Handbook of Quantitative Finance and Risk Management

“…More recently, Kalev and Duong (2008) utilize Martens and van Dijk's (2007) realized range to test the Samuelson Hypothesis for the futures contract.…”