A Survey on the Determination of Nominal Exchange Rate for USD vis-à-vis INR

In this paper, I have investigated the out of sample forecast performance for a case study on the determination of the nominal exchange rate for USD vis-à-vis IN¬R under VEC, VAR (in first difference) and Bayesian VAR specification with the help of set of economic theories. The forecast performance of these models is then compared with the random walk model, which is set as a benchmark model for forecast evaluation. The study observed that the structural model under VEC specification have superior predictive ability over BVAR and DVAR. U-statistics suggests that the yield curve model has minimum forecast error up to one year. Moreover, it is also observed that the yield curve, the tailor rule fundamental and augmented sticky price – microstructure model produces significant insight pertaining to the likely behavior in the movement of the exchange rate for a longer horizon up to 3 years through VEC specification.