The empirical findings of world and Britain’s returns in the financial market can be used to analyze the implication of the EMH model. The data used in the analysis is drawn from two populations’ the world prices and Britain’s financial market prices. The regression analysis examines if Britain’s price movements influences the world financial market prices and other statistical analysis are carried out to evaluate the statistical model.
This paper will focus on Auto correlation and serial correlation, run tests and distribution of results to draw conclusion in favor or against the theory. According to EMH past information is included in the prices thus the use of information to predict the market trends is fruitless. In addition given the fact that world market and international trade influence the economic condition of individual countries then the alternative hypothesis to the EMH would be that there is a correlation between the movement of prices in the world market and Britain’s market.
(Han, 2002 & Khan, 2004). An example is according to economic theories differential in interest rates between two countries will influence financial market trends in that domestic investor might opt to invest in foreign countries where the returns are high. If the alternative hypothesis applies then it implies data from individual’s country financial market would be serially correlated and there would be a strong correlation between Britain’s financial markets returns and world returns (Han, 2002). From the data:
• The Fama’s test shows that we are 95% sure that there is no autocorrelation between Britain’s and Worlds price movement over time. Therefore this proves that in both populations (World financial markets and Britain’s financial market. ) the error terms of different period are independent or in other words the returns of later periods do not influence current prices (Boyd, 2007). • In relations to correlation of the two population data, the results show that (both world to Britain and Britain to World correlation are not correlated at 95% confidence interval according to the fama’s test.
(Boyd, 2007). This shows that past information is not necessarily an indicator of the future trend given the fact that the data in both samples show no serial correlation between the returns in the market (Han, 2002 & Boyd, 2007). Secondly, application of the non parametric run tests which is essential in this study since it incorporate the direction of movement thus can be used to measure if there ids consistency in the movement of prices in the market.
Both populations show that we are 95% certain that in both World returns and Britain’s returns there is no evidence of clustering. This means that financial market price movement in both World returns and Britain’s returns are random further supporting the EMH theory in that, if price movement in financial market are random then there is no pattern that can be used in prediction therefore, available information and financial statement cannot information are quickly incorporated in the prices thus it is difficult to have a pattern or trend.
These findings led to the random walk theory which stipulates that prices in the financial market take a random walk. (Khan, 2004 & Wald-wolfowitz, 2001). Lastly, from the distribution of returns, the two samples several indicators support thee EMH theory: Firstly, findings from several tests show that both samples are normally distributed, the Jarques Bera test in both samples fails to reject the null hypothesis and postulate that we can be 95% sure that the samples are normally distributed.
Secondly the tests of peakedness show that the distributions take a mesokurtotic form (I. e. not too peaked nor too flat), in addition both sample are evenly skewed thus showing that the both data’s approach a perfectly normal distribution (Khan, 2004). Given that the data is composed of serial differential returns then it follows that if he EMH did not hold, the distribution would not be random but either skewed or peaked due to the fact that investor could be able to use information and make super normal profits (and losses).
Another important consideration is the mean, according to the T-test in both samples; the mean is significantly not different from zero showing that there can neither be super normal losses in the market. The fact that the data is normally distributed is consistent with the EMH theory in that price reflects all information and it is impossible to use public and private information to make abnormal profits since the information is quickly assimilated and reflected in the prices (Khan, 2004).
Boyd, D. (2007). Autocorrelation / auto-regression / serial correlation. Accessed 22nd April 2008 from University of London: Business school http://homepages. uel. ac. uk/D. A. C. Boyd/FE2007%20Autocorrelation. doc. Han, A. (2002). Efficient market hypothesis. Accessed 22nd April 2008 from http://www. alvinhan. com/Efficient-Market-Hypothesis. htm