Behavioural Biases, Demographics, Investment Strategy and Portfolio Performance of Individual Investors at the Nairobi Securities Exchange

Abstract

This study sought to establish the relationship among behavioural biases, demographics, investment strategies and portfolio performance of individual equity investors at the Nairobi Securities Exchange, Kenya. This was based on research gaps that were identified in literature, namely, conflicting results on the effect of behavioural biases on portfolio performance, contradicting evidence on whether the effect of biases differs significantly among individuals based on their demographics, and finally few studies have been done on the joint effect of behavioural biases, demographics, investment strategy on portfolio performance in Kenya. The study formulated four hypotheses so as to address the identified gaps. A sample of 400 investors was considered and questionnaires were used to collect data whereby a response rate of 69.7% was attained. Positivism research philosophy was applied to enable generalization of the findings to the whole population. The study used correlational descriptive survey design. The data was tested for reliability and validity and found appropriate for analysis. Regression diagnostics were conducted for linearity, multicollinearity and homoscedasticity. Descriptive statistics and specifically the mean, median, mode, skewness, kurtosis, and standard deviation were computed. Correlation analysis was carried out to establish the correlation among behavioural biases, demographics, investment strategy and portfolio performance. Lastly, regression analysis was carried out to test the hypotheses. The findings from the study were that there is a statistically significant relationship between behavioural biases and portfolio performance (Adjusted R2=8.1%, F=25.636, p<0.05), there is a statistically significant moderating effect of demographics on the relationship between behavioural biases and investment strategy (Adjusted R2=3.3%, F=4.191, p <0.05), there is a statistically significant intervening effect of investment strategy on the relationship between behavioural biases and portfolio performance (Adjusted R2=8.3%, F= 6.002, p<0.05) and lastly, there is a statistically significant joint effect of behavioural biases, demographics and investment strategy on portfolio performance (Adjusted R2= 10.1%, F= 5.45, p < 0.05). The findings from this study contribute to existing knowledge by firstly providing a position on the effect of behavioural biases on portfolio performance where a positive relationship was established. Secondly, by establishing the moderating effect of demographics where education was found to be a significant moderator on the relationship between behavioural biases and investment strategy. Thirdly, the mediating effect of investment strategy where contrarian strategy was found to be a significant mediator. Lastly, the joint effect of behavioural biases, demographics and investment strategy on portfolio performance, which was found to be statistically significant. This knowledge will aid agents and brokers who work with securities firms when advising investors on the appropriate investment opportunities based on their demographics. Also CMA as a regulator will use this findings to formulate policies on education programmes to be instituted so as to equip investors with knowledge on how to manage potential effects biases. The study faced a few limitations which were managed effectively by the researcher. For example some investors were unwilling to reveal information about their investments which prolonged the data collection period so as to obtain an adequate sample. The study suggests further research on other biases like local bias and ostrich effect bias as they have been scarcely studied locally. Also a study on other markets like the bond market and real assets markets to ascertain the effect of biases