It has been observed that markets are not always efficient and over the period of time many researchers have reported numerous anomalies to market efficiency that result in mispricing of securities.

The identification of profitable anomalies may be the result of a process called data mining, aka data snooping.

In a normal research practice, first of all a hypothesis is developed for the overall population and then to conclude whether the hypothesis should be accepted or rejected, tests are conducted on a sample data.

However, the data mining does exactly the reverse where the data is analyzed first with an objective of developing the hypothesis.

In the search of profitable anomalies, the data is examined in several manners, even by utilizing different empirical approaches, on the verge of detecting the desired pattern.

Although researchers can always look back on data and the excessive data snooping is likely to identify various trading strategies that might have worked in the past, yet the possibility of such strategies producing the abnormal returns just by chance cannot be denied completely.

In fact, in efficient markets, although there may appear some anomalies on frequent basis it is difficult to harness them when different factors such as risk, trading costs, etc. are taken into account.

Some of the anomalies depicting the breadth of the anomalies often observed in market are as follows: