Time-varying price discovery in Bahar-e-Azadi Gold Coin spot and futures contracts

Abstract

This paper aims to analyze the daily price discovery of Bahar-e-Azadi Gold Coin (GC) spot and futures contracts in Iran, using the fractionally cointegrated error-correction model (FCECM). The residuals of the FCECM are modeled by the BEKK-GARCH specification to calculate the time-varying conditional information share between GC spot and futures prices. Using data covering December 21, 2008 to April 14, 2018, the paper establishes the novel finding that the GC spot and futures price series are fractionally integrated of orders 0.98347 and 0.95169, respectively. This implies the long memory behavior in the price series. Further, the results show that the series are fractionally cointegrated of order 0.542. The empirical findings from the methodology indicate that in the price discovery process, the GC spot market dominates the GC futures market. This analysis is robust to alternative construction of futures price series and sub-samples decomposed based on structural breaks. One possible explanation could be the higher trading volume associated with the GC spot market compared to the GC futures market. Incompleteness and market frictions also can cause a delay in the process of information incorporation into the futures market and may discourage market players from trading in these markets.