Forecasting the Norwegian Krone exchange rate using the oil price : a trader's and a statistician's perspective

Abstract

In the past nearly one year, there has been a 30% depreciation of the Norwegian currency. The dramatic fall in the oil price is blamed to be the key reason since the oil price drop can be considered as an exogenous shock to the oil dependent Norwegian economy. If this is true, can the oil price predict movements in the Norwegian krone? We examine this issue from two different perspectives. First, we use high frequency data such as daily and hourly to simulate simple strategies involving blindly trading the dollar based on signals given by the oil price. We find that when using the direction of change in the oil price as a predictor for the direction of change in the USD/NOK exchange rate we are able earn higher risk-adjusted returns than a simple buy and hold strategy. Second, in spirit of Ferraro et al. (2015), we try to forecast daily and hourly changes in the exchange rate using oil price changes as the only predictor. We find that contemporaneous changes in the oil price significantly outperforms the random walk in terms of forecasting ability, while lagged changes in the oil price yield indistinct results highly dependent on the timing of the exchange rate data.