The lead-lag relation between index futures and the underlying index has been at the core of discussion of the researchers in the developing and the developed world. This study examines the lead-lag relationship between spot and future prices of 21 companies from the KSE-30 Index for the monthly data of the period from October 2004 to February 2008. It applies panel data estimation techniques of common constants, fixed effects and random effects. These estimations are considered to be the most recent and efficient analytical methods in handling econometric data. The data is balanced; therefore, no additional assumption has been employed. The common constants were estimated under the principle assumption of no differences among the data matrices of cross-sectional dimension. For the validity of the fixed effects, F-statistic with the assumption of homogeneity of constants against the heterogeneity proved significance of the fixed effect; notwithstanding our data of stock prices is very reasonably volatile. Yet, in order to get more satisfactory results, this study also estimates random effects. Both the GLS transformed regressions and unweighted statistics in terms of random effects are relatively better for the causal effect of future on the spot. In order to make a choice between the fixed effects and random effects, the Hausman Test was applied; which proved that the fixed effects will be always more consistent for these types of volatile data than random effects.
Lead-lag relation, Index futures, Spot market, Panel models
Journal of Economic Literature Subject Codes
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Siddiqui, M. A. (2009). Panel data models: an appraisal of lead-lag relation in the KSE-30. Business Review, 4(1), 159-176. Retrieved from https://ir.iba.edu.pk/businessreview/vol4/iss1/9
February 24, 2021