Abstract
This paper provides new evidence on the role of exchange rates in forecasting commodity prices. Consistent with previous studies, we find that commodity currencies hold out-of-sample predictive power for commodity prices when using standard linear predictive regressions. After we reconsider the evidence using noncausal autoregressions, which provide a better fit to the data and are able to accommodate the effects of nonlinearities and omitted variables, the predictive power of exchange rates disappears. (C) 2017 Elsevier B.V. All rights reserved.
Original language | English |
---|---|
Pages (from-to) | 424-433 |
Number of pages | 10 |
Journal | Energy Economics |
Volume | 65 |
DOIs | |
Publication status | Published - Jun 2017 |
MoE publication type | A1 Journal article-refereed |
Keywords
- Commodity prices
- Exchange rates
- Noncausal autoregression
- Nonlinearity
- PRESENT VALUE MODELS
- OIL PRICE SHOCKS
- EXCHANGE-RATES
- FUTURES PRICES
- MARKETS
- BUBBLES
- FUNDAMENTALS
- INFLATION
- TESTS
- BOOMS