Publication Date: November 2018
While each ﬁnancial crisis has its own characteristics, there is now widespread recognition that crises arising from sources such as ﬁnancial speculation and excessive credit creation do inflict harm on the real economy. Detecting speculative market conditions and ballooning credit risk in real time is therefore of prime importance in the complex exercises of market surveillance, risk management, and policy action. This chapter provides an R implementation of the popular real-time monitoring strategy proposed by Phillips, Shi and Yu in the International Economic Review (2015), along with a new bootstrap procedure designed to mitigate the potential impact of heteroskedasticity and to eﬀect family-wise size control in recursive testing algorithms. This methodology has been shown eﬀective for bubble and crisis detection and is now widely used by academic researchers, central bank economists, and ﬁscal regulators. We illustrate the eﬀectiveness of this procedure with applications to the S&P ﬁnancial market and the European sovereign debt sector using the psymonitor R package developed in conjunction with this chapter.
Keywords: Bubbles, Crises, Real-time detection, Recursive evolving test
JEL Classification Codes: C12, C14