Abstract:
We examine and model the performance of Initial Public Offerings (IPOs) with the advent of the Automated
Trading System (ATS) on the Efficient Market Hypothesis (EMH) of Fama (1970) and observe that the system of price
determination and encoding such information to existing and potential investors for IPOs has significantly improved with
related efficiency as most of the IPOs issued during the period after the introduction of the ATS have significantly attracted
more investor demand and commendable pricing mechanism as a result of easy and quick access to information sharing. This
could mean that information asymmetry has drastically reduced since they are electronically generated to produce the stock
prices within a very limited period of time. But until now, prices of IPOs in most cases do not fully reflect available
information as the EMH suggests and does not fulfil the Random Walk Hypothesis (Kendall, 1953, RWH) as a requirement
for weak form of market efficiency. However, despite the ATS’s immense contributions, the rate of price swings and inability
to fully reflect available information still remains an apparition to the market participants so that prices are either overpriced
or underpriced. We use the stability, stationary, and normality diagnostic tests together with the EGARCH and TGARCH to
define the trend of the prices. The result is not consistent with the Efficient Market Hypothesis of Fama (1970). Data on each
IPO daily prices were obtained from the trading statistics of Colombo Stock Exchange (CSE) consisting of 231 IPO stocks
traded between the years 2000 to 2012 consisting of 35,979 monthly observations; these prices are those of IPOs trading after
the introduction of the ATS in 1997. The outcome clearly shows that the prices are not normally distributed and are
significantly auto-correlated. This result does not support the RWH to satisfy for the weak market efficiency.
