Title: Monitoring asset price bubbles in real-time in the presence of nonstationary volatility
Authors: Matei Demetrescu - University of Kiel (Germany) [presenting]
Abstract: A real-time statistical monitoring of speculative bubbles is proposed which take into account the inherent feature of nonstationarity volatility in the innovation process of the time series of interest. Nonstationary volatility is a common but often overlooked occurrence in financial and macroeconomic time series data, such that abstracting from this process results in overdetection of bubbles. We compute critical values from a boundary value function that changes at each point in time in such a way that, asymptotically, volatility variations don't matter. We showcase proposed testing strategy in detecting real-time speculative bubbles in the valuation ratios of the S\&P 500 during the dot-com bubble as well as in bitcoin prices, yielding bubble detection rates that are robust to structural breaks.