Modelo de volatilidade estocástica com saltos aplicado a commodities agrícolas

• Autor(a) principal: Bodra, Roberto Andreotti
• Data de Publicação: 2012
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional do FGV (FGV Repositório Digital)
• Texto Completo: http://hdl.handle.net/10438/10374
• topic: Precificação; Volatilidade estocástica; Economia; Economia agrícola; Produtos agrícolas; Mercado de opções; Processo estocástico; Bolsa de mercadorias

The United States Department of Agriculture publishes every month reports with data on crop conditions, global supply and demand and inventory levels which serve as a reference for all participants in the agricultural commodities market. This market has a sharp volatility during the release of these reports. A stochastic volatility model with jumps is used to the dynamics of prices of corn and soybean. There is not an ideal model for this purpose, each one presenting its advantages and disadvantages. The chosen model was the one from Oztukel and Wilmott (1998) which is an empirical stochastic volatility model, incremented with deterministic jumps. It has been show empirically that commodities market can be well fitted under stochastic volatility models and the jump-diffusion process can properly represent the jumps that the market faces during the release of the reports. As the type of agricultural commodities options that are traded in the exchange are american, then some available methods could be used to price options under the proposed model dynamics. Given that the chosen model is multifactor, then the appropriate method for it is the one proposed by Longstaff and Schwartz (2001) called least squares Monte Carlo (LSM). Options priced by the model are then used in a strategy to hedge a physical position of corn and soybean, and the efficiency of this strategy is compared against strategies with instruments available in the market.