Analysis of Statistical Properties of Stock Price Data

This paper presents an analysis of the overreaction hypothesis in the London Stock exchange using monthly excess returns of 30 companies from the FTSE 100 index from January 2007 to January 2009. Our analysis suggests an overreaction does occur similar to the one demonstrated by Clare and Thomas (1995) who believe losses tend to be small and that overreaction can be explained by the firm size effect. However, this is not the case in our paper as we have selected 30 FTSE 100 firms.

The efficient market hypothesis states share prices fully reflect all available information, thus no possibility of profiting. However there is now a lot of evidence to suggest past prices can predict future movements in prices, resulting in abnormal profits for investors. This paper will examine the potential aspect of long run mean reversion in stock prices which is known as the overreaction hypothesis. This hypothesis is based on the evidence provided by DeBondt and Thaler which showed that stocks which performed badly over three to five year periods which are called looser would then outperform the stocks which excelled previous over the next three to five years. Using UK data from 2007 to 2009 drawn from 30 FTSE100 firms, we find losers outperform previous winners over a two year period.

This paper is constructed in the following way; in the next section we review the technique used to test for overreaction, in the third section we present our data and in the fourth section we test the overreaction hypothesis and discuss the results, the final section concludes the paper.