This paper employs a nonlinear model of bivariate generalized autoregressive conditional heteroskedasticity with mean to generate time-varying exchange-rate uncertainty and to simultaneously estimate the effects of exchange-rate uncertainty on unemployment in three developing Asian countries. This approach thus avoids overstating the level of uncertainty which has been characteristic of previous studies. It is found that the argument that lagged unemployment has an impact on exchange-rate uncertainty holds for South Korea and Taiwan, but it is less obvious in Singapore. On the other hand, in Taiwan and Singapore, the effect, that increased exchange-rate uncertainty has an obvious, positive impact on unemployment has not been substantiated.
IEEE International Conference on Grey Systems and Intelligent Services, Date of Conference: 18-20 Nov. 2007, Page(s):1426 - 1429