Project information
Attention driven market risk

Extreme market disruptions have been common in the past two decades, from the global financial crisis to the pandemic and war in Europe, all fueling demand for better risk management tools. Moreover, an oversaturated information space makes risk management harder for investors, regulators, and policymakers. The main objective of the research proposal is to quantify the extent to which investor attention drives future asset return distribution by designing and empirically verifying new market risk models—specifically, the joint estimation of Value at Risk and Expected Shortfall, and volatility. We contribute to the literature by addressing three questions. First, are we able to improve the measurement of investor's attention? Second, does it matter if we focus on overall and limited attention? Third, can we improve the forecasting accuracy of low-frequency models with machine learning and attention measures? Our research is empirical in nature based on state-of-the-art econometric and machine-learning methods.