A0947
Title: A mixture copula model for assessing net bubble values risk
Authors: Andrea Montanino - University of Naples Parthenope (Italy) [presenting]
Giovanni De Luca - University of Naples Parthenope (Italy)
Abstract: The Phillips, Shi and Yu (PSY) test is applied to investigate the presence of bubbles in the prices of cryptocurrencies and technology stocks. This methodology is based on a rolling window backward expansion procedure and is designed to allow the date stamping of the exuberant periods. The net bubble values are calculated as the difference between the Backward Supremum ADF test statistics and its critical value, estimated using a bootstrap procedure with 1000 iterations. Furthermore, the common trading days between financial assets are considered, allowing for a detailed analysis of trading patterns and synchronization among different markets. The linear correlation and tail dependence measures (through the copula function) are applied to analyze the behavior in the tails. In particular, the mixture copula is a flexible method to capture the complex dependencies in the tail regions of the distributions, which are critical to understanding the extreme co-movements and the risk of contagion. This methodology offers valuable insights for investors and policymakers, enabling them to better understand the complex dynamics of financial markets and to make informed decisions to mitigate the risks associated with explosive financial markets.
