The effect of past returns on the current trading at the Nairobi securities exchange

Abstract

Investor overconfidence has been proposed to explain various anomalous findings in security markets. The theory of investor overconfidence provides testable implication assuming investor overestimation of their abilities and private information and biased self-attribution. High (low) trading activity following market gains (losses) present on of the testable implication among others. The study sought to find out whether past returns have an effect on the trading volume at the NSE. The objective of the study was to find out how past returns influence trading activity. The population of the study was the 62 companies listed in the NSE. The companies in the 20 share index were considered as an appropriate sample for the study due to their representativeness. The weekly index and volume was obtained from the Nairobi Securities Exchange (NSE) official website and was analyzed through simple linear regression. Inconsistent with overconfidence hypothesis prediction, the findings indicated an insignificant relationship between past returns and trading volume. Based on the findings the study recommends that future studies use a longer period of time for analysis and also to analyze different sectors and indices separately. The major limitation of the study was in the method of analysis since some other variable(s)is (are) causing the variation on the dependent variable.