Rapid detection of the switching point in a financial market structure using the particle filter

We apply the particle filter for the quick and accurate estimation of a switching point in a financial market based on a recently developed theoretical model, the potentials of unbalanced complex kinetics (PUCK) model, which fulfils all empirically stylized facts such as fat-tailed distribution of price changes and the anomalous diffusion in a short-time scale. We show the efficiency of an optimized driving force in particle filtering for the estimation of the parameters of the PUCK model, using a simulation study. As an example, we apply the method to the dollar-yen exchange market before and after the biggest earthquake in Japan in March 2011. With this fast and efficient estimation method, we can clearly confirm that the statistics of the time series of exchange rate changed drastically at the time of the arrival of the quake in Tokyo area, implying that the earthquake worked as a trigger for the market's switching point.