Forecasting Jet Fuel Prices Using Artificial Neural Networks

Artificial neural networks provide a new approach to commodity forecasting that does not require algorithm or rule development. Neural networks have been deemed successful in applications involving optimization, classification, identification, pattern recognition and time series forecasting. With the advent of user friendly, commercially available software packages that work in a spreadsheet environment, such as Neural Works Predict by NeuralWare, more people can take advantage of the power of artificial neural networks. This thesis provides an introduction to neural networks, and reviews two recent studies of forecasting commodities prices. This study also develops a neural network model using Neural Works Predict that forecasts jet fuel prices for the Defense Fuel Supply Center (DFSC). In addition, the results developed are compared to the output of an econometric regression model, specifically, the Department of Energy's Short-Term Integrated Forecasting System (STWS) model. The Predict artificial neural network model produced more accurate results and reduced the contribution of outliers more effectively than the STIFS model, thus producing a more robust model.

Artificial neural networks provide a new approach to commodity forecasting that does not require algorithm or rule development. Neural networks have been deemed successful in applications involving optimization, classification, identification, pattern recognition and time series forecasting. With the advent of user friendly, commercially available software packages that work in a spreadsheet environment, such as Neural Works Predict by NeuralWare, more people can take advantage of the power of artificial neural networks. This thesis provides an introduction to neural networks, and reviews two recent studies of forecasting commodities prices. This study also develops a neural network model using Neural Works Predict that forecasts jet fuel prices for the Defense Fuel Supply Center (DFSC). In addition, the results developed are compared to the output of an econometric regression model, specifically, the Department of Energy's Short-Term Integrated Forecasting System (STWS) model. The Predict artificial neural network model produced more accurate results and reduced the contribution of outliers more effectively than the STIFS model, thus producing a more robust model.

This paper applies a recurrent neural network (RNN) method to forecast cotton and oil prices. We show how these new tools from machine learning, particularly Long-Short Term Memory (LSTM) models, complement traditional methods. Our results show that machine learning methods fit reasonably well with the data but do not outperform systematically classical methods such as Autoregressive Integrated Moving Average (ARIMA) or the naïve models in terms of out of sample forecasts. However, averaging the forecasts from the two type of models provide better results compared to either method. Compared to the ARIMA and the LSTM, the Root Mean Squared Error (RMSE) of the average forecast was 0.21 and 21.49 percent lower, respectively, for cotton. For oil, the forecast averaging does not provide improvements in terms of RMSE. We suggest using a forecast averaging method and extending our analysis to a wide range of commodity prices.

Masters Theses in the Pure and Applied Sciences was first conceived, published, and disseminated by the Center for Information and Numerical Data Analysis and Synthesis (CINDAS)* at Purdue University in 1957, starting its coverage of theses with the academic year 1955. Beginning with Volume 13, the printing and dis semination phases of the activity were transferred to University Microfilms/Xerox of Ann Arbor, Michigan, with the thought that such an arrangement would be more beneficial to the academic and general scientific and technical community. After five years of this jOint undertaking we had concluded that it was in the interest of all concerned if the printing and distribution of the volumes were handled by an international publishing house to assure improved service and broader dissemination. Hence, starting with Volume 18, Masters Theses in the Pure and Applied Sciences has been disseminated on a worldwide basis by Plenum Publishing Corporation of New York, and in the same year the coverage was broadened to include Canadian universities. All back issues can also be ordered from Plenum. We have reported in Volume 40 (thesis year 1995) a total of 10,746 thesis titles from 19 Canadian and 144 United States universities. We are sure that this broader base for these titles reported will greatly enhance the value of this impor tant annual reference work. While Volume 40 reports theses submitted in 1995, on occasion, certain uni versities do report theses submitted in previous years but not reported at the time.

The aim of this publication is to identify and apply suitable methods for analysing and predicting the time series of gold prices, together with acquainting the reader with the history and characteristics of the methods and with the time series issues in general. Both statistical and econometric methods, and especially artificial intelligence methods, are used in the case studies. The publication presents both traditional and innovative methods on the theoretical level, always accompanied by a case study, i.e. their specific use in practice. Furthermore, a comprehensive comparative analysis of the individual methods is provided. The book is intended for readers from the ranks of academic staff, students of universities of economics, but also the scientists and practitioners dealing with the time series prediction. From the point of view of practical application, it could provide useful information for speculators and traders on financial markets, especially the commodity markets.

Processing information and analyzing data efficiently and effectively is crucial for any company that wishes to stay competitive in its respective market. Nonlinear data presents new challenges to organizations, however, due to its complexity and unpredictability. The only technology that can properly handle this form of data is artificial neural networks. These modeling systems present a high level of benefits in analyzing complex data in a proficient manner, yet considerable research on the specific applications of these intelligent components is significantly deficient. Applications of Artificial Neural Networks for Nonlinear Data is a collection of innovative research on the contemporary nature of artificial neural networks and their specific implementations within data analysis. While highlighting topics including propagation functions, optimization techniques, and learning methodologies, this book is ideally designed for researchers, statisticians, academicians, developers, scientists, practitioners, students, and educators seeking current research on the use of artificial neural networks in diagnosing and solving nonparametric problems.

Artificial neural networks (ANNs) present many benefits in analyzing complex data in a proficient manner. As an effective and efficient problem-solving method, ANNs are incredibly useful in many different fields. From education to medicine and banking to engineering, artificial neural networks are a growing phenomenon as more realize the plethora of uses and benefits they provide. Due to their complexity, it is vital for researchers to understand ANN capabilities in various fields. The Research Anthology on Artificial Neural Network Applications covers critical topics related to artificial neural networks and their multitude of applications in a number of diverse areas including medicine, finance, operations research, business, social media, security, and more. Covering everything from the applications and uses of artificial neural networks to deep learning and non-linear problems, this book is ideal for computer scientists, IT specialists, data scientists, technologists, business owners, engineers, government agencies, researchers, academicians, and students, as well as anyone who is interested in learning more about how artificial neural networks can be used across a wide range of fields.

A fundamental objective of Artificial Intelligence (AI) is the creation of in telligent computer programs. In more modest terms AI is simply con cerned with expanding the repertoire of computer applications into new domains and to new levels of efficiency. The motivation for this effort comes from many sources. At a practical level there is always a demand for achieving things in more efficient ways. Equally, there is the technical challenge of building programs that allow a machine to do something a machine has never done before. Both of these desires are contained within AI and both provide the inspirational force behind its development. In terms of satisfying both of these desires there can be no better example than machine learning. Machines that can learn have an in-built effi ciency. The same software can be applied in many applications and in many circumstances. The machine can adapt its behaviour so as to meet the demands of new, or changing, environments without the need for costly re-programming. In addition, a machine that can learn can be ap plied in new domains with the genuine potential for innovation. In this sense a machine that can learn can be applied in areas where little is known about possible causal relationships, and even in circumstances where causal relationships are judged not to exist. This last aspect is of major significance when considering machine learning as applied to fi nancial forecasting.