Stock market predictability: Non-synchronous trading or inefficient markets? Evidence from the National Stock Exchange of India

Purpose: The main objective of this study is to obtain new empirical evidence on non-synchronous trading effects through modelling the predictability of market indices. Design / Methodology / Approach: We test for lead-lag effects between the Indian Nifty and Nifty Junior indices using Pesaran-Timmermann tests and Granger-Causality. We then propose a simple test on overnight returns, in order to infer whether the observed predictability is mainly attributable to non-synchronous trading or some form of inefficiency. Findings: The evidence suggests that non-synchronous trading is a better explanation for the observed predictability in the Indian stock market. Research limitations / implications: The indication that non-synchronous trading effects become more pronounced in high-frequency data, suggests that prior studies using daily data may underestimate the impacts of non-synchronicity. Originality / value: The originality of the paper rests on various important contributions: (a) we look at overnight returns to infer whether predictability is more attributable to non-synchronous trading or to some form of inefficiency, (b) we investigate the impacts of non-synchronicity in terms of lead-lag effects rather than serial correlation, and (c) we use high-frequency data which gauges the impacts of non-synchronicity during less active parts of the trading day.