Application of a Modified TAR Model to CIP Deviations in Asian Data

The methodology to be used in this paper is estimation of a threshold autoregressive (TAR) model. In this model deviations are random within a band defined by transactions costs and contract risk, and autoregressive towards the band outside it. The principal reference is Tchernykh (1998). These estimates can provide indicators for policy-makers of the market¡¦s expectation of crisis. They could also provide indicators for the private sector of convergence of deviations to their usual bands. The estimation methodology is a non-linear three-regime maximum likelihood procedure. The TAR model has the potential to be applied to differentials between linked pairs of financial market prices more generally. This paper modifies the classical TAR model to allow for progressive deviations from a stochastic regime, rather than simple jumps.