A Non Parametric Model for the Forecasting of the Venezuelan Oil Prices

TL;DR

This paper introduces a neural network model for forecasting Venezuelan oil prices, utilizing inputs based on a dynamic system model, and evaluates its predictive performance through statistical tests and trading strategy analysis.

Contribution

The paper presents a non-parametric neural network approach for oil price forecasting using dynamically selected inputs, with performance evaluation based on profitability tests and trading strategy characteristics.

Findings

The neural network model effectively predicts oil prices.

The model's predictions enable profitable trading strategies.

Statistical tests confirm the model's predictive power.

Abstract

A neural net model for forecasting the prices of Venezuelan crude oil is proposed. The inputs of the neural net are selected by reference to a dynamic system model of oil prices by Mashayekhi (1995, 2001) and its performance is evaluated using two criteria: the Excess Profitability test by Anatoliev and Gerko (2005) and the characteristics of the equity curve generated by a trading strategy based on the neural net predictions. ----- Se introduce aqui un modelo no parametrico para pronosticar los precios del petroleo Venezolano cuyos insumos son seleccionados en base a un sistema dinamico que explica los precios en terminos de dichos insumos. Se describe el proceso de recoleccion y pre-procesamiento de datos y la corrida de la red y se evaluan sus pronosticos a traves de un test estadistico de predictibilidad y de las caracteristicas del Equity Curve inducido por la estrategia de compraventa bursatil generada por dichos pronosticos.