Nonlinear deterministic forecasting of daily Peseta–Dollar exchange rate

We perform out-of-sample predictions on daily Peseta–Dollar spot exchange rates using a simple nonlinear deterministic technique of local linear predictor. We compared our predictions with those by two simple benchmark predictors: random walk model and mean-value predictor. The results on the differenced time series indicate that our predictions are better than those by the random walk model, and marginally better than the results from the mean-value predictor. .

Time series; Embedding dimension; Nonlinear prediction; Exchange rate; Random walk