Skip to content
Back to outputs

Oil price volatility is effective in predicting food price volatility. Or is it?

Research output: Contribution to journal › Article › peer-review

Standard

Oil price volatility is effective in predicting food price volatility. Or is it? / Chatziantoniou, Ioannis; Degiannakis, Stavros; Filis, George; Lloyd, Tim.

In: Energy Journal, 09.10.2020.

Research output: Contribution to journal › Article › peer-review

Harvard

Chatziantoniou, I, Degiannakis, S, Filis, G & Lloyd, T 2020, 'Oil price volatility is effective in predicting food price volatility. Or is it?', Energy Journal.

APA

Chatziantoniou, I., Degiannakis, S., Filis, G., & Lloyd, T. (Accepted/In press). Oil price volatility is effective in predicting food price volatility. Or is it? Energy Journal.

Vancouver

Chatziantoniou I, Degiannakis S, Filis G, Lloyd T. Oil price volatility is effective in predicting food price volatility. Or is it? Energy Journal. 2020 Oct 9.

Author

Chatziantoniou, Ioannis ; Degiannakis, Stavros ; Filis, George ; Lloyd, Tim. / Oil price volatility is effective in predicting food price volatility. Or is it?. In: Energy Journal. 2020.

Bibtex

@article{94fc512d5ff04d44b2991e12e9fbe420,
title = "Oil price volatility is effective in predicting food price volatility. Or is it?",
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.",
keywords = "Forecasting, Food price volatility, Heterogeneous Autoregressive, Mixed-data sampling, Oil price volatility, Model Confidence Set, embargoover12",
author = "Ioannis Chatziantoniou and Stavros Degiannakis and George Filis and Tim Lloyd",
note = "36 month embargo. ",
year = "2020",
month = oct,
day = "9",
language = "English",
journal = "The Energy Journal",
issn = "0195-6574",
publisher = "International Association for Energy Economics",

}

RIS

TY - JOUR

T1 - Oil price volatility is effective in predicting food price volatility. Or is it?

AU - Chatziantoniou, Ioannis

AU - Degiannakis, Stavros

AU - Filis, George

AU - Lloyd, Tim

N1 - 36 month embargo.

PY - 2020/10/9

Y1 - 2020/10/9

N2 - 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.

AB - 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.

KW - Forecasting

KW - Food price volatility

KW - Heterogeneous Autoregressive

KW - Mixed-data sampling

KW - Oil price volatility

KW - Model Confidence Set

KW - embargoover12

M3 - Article

JO - The Energy Journal

JF - The Energy Journal

SN - 0195-6574

ER -

ID: 22978764