An Empirical investigation of the relationship between investor sentiment and volatility of the equities market in Kenya between 2013 and 2023

Abstract

Volatility is a key component of financial markets. It has direct bearing on market uncertainty and influences investment decisions. Higher stock market return volatility can enhance liquidity, but excessive volatility may precipitate crashes. Stock market volatility in Kenya has exhibited an upward trend which raises concerns on heightened market risk. This underscores the need for a comprehensive understanding of the factors driving volatility in the Kenyan market, including investor sentiment, which can cause price swings that are not always explained by fundamentals. This research therefore sought to examine the relationship between investor sentiment and volatility in the Kenyan stock market, guided by the efficient market hypothesis, market microstructure theory, and the noise trader theory. The study period spanned from 2013 to 2023 capturing significant events such as the 2013, 2017 and 2022 elections, the Covid-19 pandemic, currency depreciation and other socio-political-economic shifts which could impact the sentiment-volatility relationship. The study aimed at capturing and quantifying the prevailing market wide sentiment at various time intervals and on further analysis of the sentiment-volatility relationship. The asymmetric effect of sentiment on volatility and the variation in the relationship between high and low volatility regimes was also assessed. Investor sentiment was measured using a monthly sentiment index created using principal component analysis from market-based indicators, while controlling for macroeconomic variables and company fundamentals. The study leveraged on the Generalised Autoregressive Conditional Heteroscedasticity models and quantile regression to capture fluctuations in volatility over time and the impact of investor sentiment. The findings indicated that the sentiment index has significant impact on the return volatility in the Kenyan stock market, with positive sentiment increasing and negative sentiment decreasing volatility. This relationship varies across volatility regimes, being negative during low-volatility periods and positive during average and high-volatility periods. These findings offer practical implications for integrating sentiment analysis into investment and risk management strategies. Limitations include reliance on proxy-based sentiment measures and focus on a single market, suggesting avenues for future research using psychological sentiment indices and multi-market comparisons.