tail behavior of a threshold autoregressive stochastic volatility model

International audienceWe consider a threshold autoregressive stochastic volatility model where the driving noises are sequences of iid regurlarly random vatiables. We prove that both the right and the left tails of the marginal distribution of the log-volatility process are regularly varying with tail exponent. We also determine the exact values of the coefficients in the tail of the considered process