Market Efficiency Tests: Five Anomalies Exist in Both Us and Australian Markets

Only available on StudyMode
  • Download(s) : 149
  • Published : August 20, 2008
Open Document
Text Preview
Executive summary
Market efficiency tests include weak, semi-strong and strong three forms. They assume that financial markets are "informationally efficient", or that prices of trading assets, already reflect all known information and therefore are unbiased in the sense that they reflect the collective beliefs of all investors about future prospects. The weak form test is based on the past information and public available information for semi-strong while strong form covers not only the public but also the private information.

Five market anomalies that appeared in U.S. and Australian equity market are discussed. They are Book-to-market ratio, January effect, small firm effect, weekend and temperature effect. The small firm effect seems to be disappeared in the US market in recent years and existed in Australia until 1994. January effect is vanishing in both US and Australian market. There is a statistically significant negative correlation between stock returns and temperature is found in both U.S and Australian equity market and theory of stock returns and temperature turned out to be true. In contrast to US market, Australian stock markets appeared to exhibit a Tuesday effect rather than Monday effect (Weekend effect). For the book-to-market ratio, there is significant evidence that the higher book-to-market ratios are related to higher returns in U.S whereas the book-to-market ratios are not significantly related to return in Australian.

The active fund managers always earn short term excess returns, but they can not outperform a passive buy-and–hold (BHM) investment strategy for US equity mutual funds with wavy returns in the future. In Australian market, the active fund managers are now less capable of outperforming BHM investment because the performance has been less significant for the last decade and there are several biases that limits in assessing whether the active managers can outperform the market.

As a result of the report, the markets are not always efficient in both US and Australia.

Weak form test
The weak form hypothesis states that all information contained in past price, volume or short interest movements is fully reflected in current market prices. It is called weak form because the security prices are the most public and easily accessible to the information. It implies that no one is able to perform well in the market using information that is known to everyone. This technique is known as technical analysis states it is ineffective for predicting future price changes. A serial correlation is a method of measuring stock market returns by sensitive trends in stock prices. Positive serial correlation is when positive returns are likely to follow positive return, which is also known as momentum effect. Momentum effect refers to using past returns in one year of less, and buying stocks that have performed well in the past and selling stocked that performed poorly in the past. The studies by Conrad and Kaul and Lo and MacKinlay found that positive serial correlation occurs in the short horizons of weekly returns in NYSE stocks. Because the correlation coefficients of weekly returns were small, the studies do not clearly suggest the existence of trading opportunities. Jagadessh and Titman examined behaviour of stock price of intermediate-horizon (3 to 12 months). They discovered that momentum strategy exists where best-performing stocks tend to performance similarly over time. They concluded that momentum strategy is profitable in an intermediate-horizon. However, there is a problem when buying and selling the stock in the short run. It requires a large amount of brokerage due to frequent buying and selling in the short run. Long horizon return gives opposite results to the short-term returns. In contrast to short-term returns, there is some evidence that returns tend to revert towards the mean, hence negative long-term serial correlation happens. A...
tracking img