Individual mean-variance relation and stock-level investor sentiment

Abstract

This research studies the effect of stock-level investor sentiment on individual stock returns’ mean-variance relation. Using unique buy and sell volume data of retail investors in Korean stock market, we find that a positive mean-variance relation is undermined among high-sentiment stocks, but holds among low-sentiment stocks. We adopt buy-sell imbalances of retail investors for individual stocks as a measure of stock-level investor sentiment. Further, our findings provide empirical evidence of a strong risk-return trade-off among stocks with low retail concentration (e.g., large capitalization, high-priced, and growth stocks). Existing research only analyzes market-wide investor sentiment. However, we study the effect of stock-level investor sentiment on individual stock returns. Therefore, our findings suggest novel implications about the investment strategy that the stock-level investor sentiment is important when constructing portfolios based on variance.

Keywords:

• investor sentiment
• mean-variance relation
• risk-return trade-off
• conditional variance
• buy-sell imbalance
• individual stock markets

JEL Classification:

• G12
• G14
• G15

Additional information

Notes on contributors

Jun Sik Kim

Jun Sik KIM is an Assistant Professor in the Division of International Trade at Incheon National University. His research interests cover the areas of volatility forecasting, behavioral finance, and empirical asset pricing.

Da-Hea Kim

Da-Hea KIM is an Assistant Professor in SKK Business School, Sungkyunkwan University. Her research interests cover the areas of empirical asset pricing, derivatives, and behavioral finance.

Sung Won Seo

Sung Won SEO is an Assistant Professor in School of Business at Ajou University. He is interested in the areas of behavioral finance, empirical asset pricing, and corporate finance.