Oil price volatility is effective in predicting food price volatility. Or is it?
Authors: Chatziantoniou, I., Degiannakis, S., Filis, G. and Lloyd, T.
Journal: Energy Journal
Volume: 42
Issue: 6
Pages: 1-24
ISSN: 0195-6574
DOI: 10.5547/01956574.42.6.ICHA
Abstract:Volatility spillovers between food commodities and oil prices have been identified in the literature, yet, there has been no empirical evidence to suggest that oil price volatility improves real out-of-sample forecasts of food price volatility. In this study we provide new evidence showing that oil price volatility does not improve forecasts of agricultural price volatility. This finding is based on extensive and rigorous testing of five internationally traded agricultural commodities (soybeans, corn, sugar, rough rice and wheat) and two oil benchmarks (Brent and WTI). We employ monthly and daily oil and food price volatility data and two forecasting frameworks, namely, the HAR and MIDAS-HAR, for the period 2nd January 1990 until 31st March 2017. Results indicate that oil volatility-enhanced HAR or MIDAS-HAR models cannot systematically outperform the standard HAR model. Thus, contrary to what has been suggested by the existing literature based on in-sample analysis, we are unable to find any systematic evidence that oil price volatility improves out-of-sample forecasts of food price volatility. The results remain robust to the choice of different out-of-sample forecasting periods and three different volatility measures.
https://eprints.bournemouth.ac.uk/34993/
Source: Scopus
Oil price volatility is effective in predicting food price volatility. Or is it?
Authors: Chatziantoniou, I., Degiannakis, S., Filis, G. and Lloyd, T.
Journal: Energy Journal
Volume: 42
Issue: 6
Publisher: International Association for Energy Economics
ISSN: 0195-6574
DOI: 10.5547/01956574.42.6.icha
Abstract:Volatility spillovers between food commodities and oil prices have been identified in the literature, yet, there has been no empirical evidence to suggest that oil price volatility improves real out-of-sample forecasts of food price volatility. In this study we provide new evidence showing that oil price volatility does not improve forecasts of agricultural price volatility. This finding is based on extensive and rigorous testing of five internationally traded agricultural commodities (soybeans, corn, sugar, rough rice and wheat) and two oil benchmarks (Brent and WTI). We employ monthly and daily oil and food price volatility data and two forecasting frameworks, namely, the HAR and MIDAS-HAR, for the period 2nd January 1990 until 31st March 2017. Results indicate that oil volatility-enhanced HAR or MIDAS-HAR models cannot systematically outperform the standard HAR model. Thus, contrary to what has been suggested by the existing literature based on in-sample analysis, we are unable to find any systematic evidence that oil price volatility improves out-of-sample forecasts of food price volatility. The results remain robust to the choice of different out-of-sample forecasting periods and three different volatility measures.
https://eprints.bournemouth.ac.uk/34993/
Source: Manual
Oil price volatility is effective in predicting food price volatility. Or is it?
Authors: Chatziantoniou, I., Degiannakis, S., Filis, G. and Lloyd, T.
Journal: Energy Journal
Volume: 42
Issue: 6
ISSN: 0195-6574
Abstract:Volatility spillovers between food commodities and oil prices have been identified in the literature, yet, there has been no empirical evidence to suggest that oil price volatility improves real out-of-sample forecasts of food price volatility. In this study we provide new evidence showing that oil price volatility does not improve forecasts of agricultural price volatility. This finding is based on extensive and rigorous testing of five internationally traded agricultural commodities (soybeans, corn, sugar, rough rice and wheat) and two oil benchmarks (Brent and WTI). We employ monthly and daily oil and food price volatility data and two forecasting frameworks, namely, the HAR and MIDAS-HAR, for the period 2nd January 1990 until 31st March 2017. Results indicate that oil volatility-enhanced HAR or MIDAS-HAR models cannot systematically outperform the standard HAR model. Thus, contrary to what has been suggested by the existing literature based on in-sample analysis, we are unable to find any systematic evidence that oil price volatility improves out-of-sample forecasts of food price volatility. The results remain robust to the choice of different out-of-sample forecasting periods and three different volatility measures.
https://eprints.bournemouth.ac.uk/34993/
http://www.iaee.org/en/publications/ejarticle.aspx?id=3751
Source: BURO EPrints