The Random-Walk Hypothesis on the Indian Stock Market

Abstract

We test the random-walk hypothesis for the Indian stock market by applying three unit root tests with two structural breaks. We find that unit root tests that allow for two structural breaks alone are not able to reject the unit root null; however, a recently developed unit root test that simultaneously accounts for heteroskedasticity and structural breaks finds that the stock indexes are mean reverting. Our results point to the importance of addressing heteroskedasticity when testing for a random walk with high-frequency financial data.

KEY WORDS:

• India
• structural break
• stock market
• random walk
• unit root

Acknowledgement

We thank Paresh Narayan for providing the GAUSS codes for the Narayan and Liu (2013) and Narayan and Popp (2010) unit root tests.

ORCID

Vinod Mishra

http://orcid.org/0000-0002-4686-5195

Notes

^1. Other applications of the Narayan and Liu (2013) test are Mishra and Smyth (2014) (United States natural gas consumption), Narayan and Liu (2011) (commodity prices), and Salisu and Mobolaji (2013) (exchange rates and oil prices). With the exception of Mishra and Smyth (2014), these studies are based on earlier working paper versions of Narayan and Liu (2013).

^2. For details on the sequential procedure approach, refer to Narayan and Popp (2010).

^3. The main reason for the crash was that the stockbroker Ketan Parekh (KP) and his allies bought large stakes in technology stocks and kept on carrying forward the payments, using securities as collateral. With the crash of technology stocks in the United States, the price of technology stocks dropped drastically and KP failed to pay back his loans (or provide more collateral) and defaulted on his loans made for stock purchases. The system that allowed the rollover of security purchase settlements was popularly known as the badla system.

^4. The old badla system combined features of forward and margin trading. It was legal but minimally regulated, and this risky system allowed investors to trade stocks with little cash. The investors were allowed to settle the trade up to five days later and even pay a fee to delay settlement still longer. This system had features of derivatives trading, but instead of exchange being an intermediary between buyer and seller, the stock brokers acted as intermediaries.