EcoSta 2017: Start Registration
View Submission - EcoSta2017
A0627
Title: The count of Monte Carlo: Analysing global banking crises, 1800-2010 Authors:  Andrew Tremayne - University of Liverpool (United Kingdom) [presenting]
Mardi Dungey - University of Tasmania (Australia)
Vance Martin - University of Melbourne (Australia)
Chrismin Tang - University of Melbourne (Australia)
Abstract: A new class of models is proposed to capture the dynamic transmission mechanisms of banking crises across national borders. As the data are characterized as counts of the number of countries experiencing a banking crisis within a year, a threshold integer autoregressive moving average model (TINARMA) is specified, where the threshold allows for the decomposition into systemic and nonsystemic crisis periods. As maximum likelihood methods are infeasible to estimate the models parameters, a simulation based procedure using efficient method of moments is adopted with standard errors based on subsampling. Applying the modelling framework using annual data from 1800 to 2010, shows that the model is able to capture the dynamic transmission mechanisms of banking crises over time. The forecasting properties of the model are also explored with a focus on periods of financial stress.