Excessive food price variability early warning system: Incorporating exogenous covariates

The Excessive Food Price Variability Early Warning System, maintained by IFPRI’s Food Security Portal, identifies periods of unusual or excessive price volatility (i.e. price variability that exceeds a pre-established, estimated threshold) for a wide set of major agricultural and food commodity mark...

Descripción completa

Detalles Bibliográficos
Autores principales: Geng, Xin, Hernandez, Manuel A., Martins-Filho, Carlos
Formato: Artículo preliminar
Lenguaje:Inglés
Publicado: International Food Policy Research Institute 2021
Materias:
Acceso en línea:https://hdl.handle.net/10568/143600
_version_ 1855540888303828992
author Geng, Xin
Hernandez, Manuel A.
Martins-Filho, Carlos
author_browse Geng, Xin
Hernandez, Manuel A.
Martins-Filho, Carlos
author_facet Geng, Xin
Hernandez, Manuel A.
Martins-Filho, Carlos
author_sort Geng, Xin
collection Repository of Agricultural Research Outputs (CGSpace)
description The Excessive Food Price Variability Early Warning System, maintained by IFPRI’s Food Security Portal, identifies periods of unusual or excessive price volatility (i.e. price variability that exceeds a pre-established, estimated threshold) for a wide set of major agricultural and food commodity markets. The Tool is based on nonparametric estimators for conditional value-at-risk (CVaR) and conditional expected shortfall (CES) associated with conditional distributions of the modeled price returns series, developed by Martins-Filho et al. (2015, 2018). The estimations are performed on a day-to-day basis and intend to provide timely price variability alerts to makers, traders, and farmers worldwide. This incorporates exogenous covariates into the Tool when modeling the stochastic behavior of commodity price returns. We focus on maize, wheat, and soybeans, which are key agricultural commodities, and add three exogenous variables when estimating the conditional mean and variance of their daily price returns. The exogeneous covariates include oil price returns, exchange rate percentage variations (US dollar depreciation rate), and Standard & Poor index (S&P500) returns, which are available on a daily frequency and are widely used when modeling dynamic price relationships in agricultural commodity markets (see, e.g., Deb et al., 1996; Ai et al., 2006; Gilbert, 2010; Natanelov, 2011; Gardebroek et al., 2016). The objective of the is twofold. First, we assess whether the behavior of agricultural commodity prices can be associated with energy (input), macroeconomic, or financial factors by depicting the relationships between variations in specific exogenous covariates and agricultural price returns and volatility. Second, we evaluate whether incorporating these covariates can help improve the predictive performance of the Tool in terms of more accurately identifying periods of excessive volatility in agricultural markets, relative to a model without exogenous variables that only includes lagged own agricultural price returns as covariates. The period of analysis is January 1990 through December 2020.
format Artículo preliminar
id CGSpace143600
institution CGIAR Consortium
language Inglés
publishDate 2021
publishDateRange 2021
publishDateSort 2021
publisher International Food Policy Research Institute
publisherStr International Food Policy Research Institute
record_format dspace
spelling CGSpace1436002025-12-08T10:11:39Z Excessive food price variability early warning system: Incorporating exogenous covariates Geng, Xin Hernandez, Manuel A. Martins-Filho, Carlos models maize soybeans food security price volatility food prices wheat The Excessive Food Price Variability Early Warning System, maintained by IFPRI’s Food Security Portal, identifies periods of unusual or excessive price volatility (i.e. price variability that exceeds a pre-established, estimated threshold) for a wide set of major agricultural and food commodity markets. The Tool is based on nonparametric estimators for conditional value-at-risk (CVaR) and conditional expected shortfall (CES) associated with conditional distributions of the modeled price returns series, developed by Martins-Filho et al. (2015, 2018). The estimations are performed on a day-to-day basis and intend to provide timely price variability alerts to makers, traders, and farmers worldwide. This incorporates exogenous covariates into the Tool when modeling the stochastic behavior of commodity price returns. We focus on maize, wheat, and soybeans, which are key agricultural commodities, and add three exogenous variables when estimating the conditional mean and variance of their daily price returns. The exogeneous covariates include oil price returns, exchange rate percentage variations (US dollar depreciation rate), and Standard & Poor index (S&P500) returns, which are available on a daily frequency and are widely used when modeling dynamic price relationships in agricultural commodity markets (see, e.g., Deb et al., 1996; Ai et al., 2006; Gilbert, 2010; Natanelov, 2011; Gardebroek et al., 2016). The objective of the is twofold. First, we assess whether the behavior of agricultural commodity prices can be associated with energy (input), macroeconomic, or financial factors by depicting the relationships between variations in specific exogenous covariates and agricultural price returns and volatility. Second, we evaluate whether incorporating these covariates can help improve the predictive performance of the Tool in terms of more accurately identifying periods of excessive volatility in agricultural markets, relative to a model without exogenous variables that only includes lagged own agricultural price returns as covariates. The period of analysis is January 1990 through December 2020. 2021-09-27 2024-05-22T12:15:30Z 2024-05-22T12:15:30Z Working Paper https://hdl.handle.net/10568/143600 en https://hdl.handle.net/10568/154182 Open Access application/pdf International Food Policy Research Institute Geng, Xin; Hernandez, Manuel A.; and Martins-Filho, Carlos. 2021. Excessive food price variability early warning system: Incorporating exogenous covariates. IFPRI Technical September 2021. Washington, DC: International Food Policy Research Institute (IFPRI). https://doi.org/10.2499/p15738coll2.134592>.
spellingShingle models
maize
soybeans
food security
price volatility
food prices
wheat
Geng, Xin
Hernandez, Manuel A.
Martins-Filho, Carlos
Excessive food price variability early warning system: Incorporating exogenous covariates
title Excessive food price variability early warning system: Incorporating exogenous covariates
title_full Excessive food price variability early warning system: Incorporating exogenous covariates
title_fullStr Excessive food price variability early warning system: Incorporating exogenous covariates
title_full_unstemmed Excessive food price variability early warning system: Incorporating exogenous covariates
title_short Excessive food price variability early warning system: Incorporating exogenous covariates
title_sort excessive food price variability early warning system incorporating exogenous covariates
topic models
maize
soybeans
food security
price volatility
food prices
wheat
url https://hdl.handle.net/10568/143600
work_keys_str_mv AT gengxin excessivefoodpricevariabilityearlywarningsystemincorporatingexogenouscovariates
AT hernandezmanuela excessivefoodpricevariabilityearlywarningsystemincorporatingexogenouscovariates
AT martinsfilhocarlos excessivefoodpricevariabilityearlywarningsystemincorporatingexogenouscovariates