Testing of Consistent Trends in Stock Performance in The Nairobi Securities Exchange

Abstract

Stocks represent units of capital required to finance operations of firms. Stock markets contain listed companies whose prices adjust randomly based on release of corporate news which also arises randomly. The instantaneous incorporation of news in stock prices through competition among market participants makes stock markets efficient. Consistent stock performance contradicts random adjustment of stock prices in efficient markets and is anomalous although has the potential of generating significant profits. This research set out to do the following: to test the existence of consistent stock performance in the NSE during the years 2001 to 2010, to examine whether consistent stock performance is associated with efficiency of NSE, to assess whether consistent stock performance in the NSE is related to: market anomalies, stock value and underlying firm characteristics. The research employed balanced monthly closing average stock price data for 32 sample stocks drawn using purposive sampling technique from a population of 56 stocks listed in the NSE during the study period. In order to identify consistent stock performance, frequency tests were employed. In order to test association between consistent stock performance and efficiency of NSE t-tests were employed for significance of abnormal returns, Spearman rank correlation was employed for volatility of stock prices with and Runs test was employed for serial correlation of stock returns. The relationship between consistent stock performance and stock market anomalies in NSE was tested using simple regression analysis for size and calendar anomalies. Spearman rank correlation test was employed for overreaction anomaly. The relationship between consistent stock performance and stock value was tested using simple regression analysis. The relationship between consistent stock performance and underlying firm characteristics was tested using multi-regression analysis. The results indicated that there was weak evidence of consistent stock performance in the NSE. There was insignificant relationship between consistent stock performance and weak form efficiency of NSE. There was insignificant relationship between consistent stock performance and size anomaly in NSE. However the market was not free from calendar and under reaction anomalies. There was an insignificant relationship between consistent stock performance and si:ock value. There was insignificant relationship between consistent stock performance and underlying firm characteristics. The overall results was insignificant presence of consistent stock performance in the NSE which confirms that the market may be weak form efficient. This research has contributed to knowledge by enhancing the threshold of determining the returns consistent winner and loser stocks which are now required to exceed 10% on a daily basis. The alternative definitions of consistent stock performance have been studied in the past independently but in this research the definitions have been combined to mitigate the inherent weaknesses that the definitions independently possess. Proportional runs volatility metric has been innovated in this research from requirements volatility metric of Computer Science discipline as the new model is successfully able to measure stock return volatility.