Elsevier, Neural Networks In Finance 2005_1 pdf

[...]... Conference on Neural Networks (IJCNN) meetings in Washington, DC, in 2001, and in Honolulu and Singapore in 2002 These meetings were eye-openers for anyone trained in classical statistics and econometrics and illustrated the breadth of applications of neural network research I wish to thank my fellow Jesuits at Georgetown University and in Washington, DC, who have been my “company” since my arrival... forecasting, and control in volatile, increasingly complex, multidimensional environments Researchers need no longer confine themselves to linear or log-linear models, or assume that underlying stochastic processes are Gaussian or normal in order to obtain forecasts and pinpoint risk–return trade-offs In short, we can go beyond linearity and normality in our assumptions with the use of neural networks. .. making, in the light of increasing financial market volatility and internationalized capital flows, cannot be overexaggerated The past two decades have witnessed extreme macroeconomic instability, first in Latin America and then in Asia Thus, both financial analysts and decision makers cannot help but be interested in predicting the underlying rates of return and spreads, as well as the default rates, in. .. forecasts guiding the decision makers work Diagnostic and forecasting methods feed back in very direct ways to decision-making environments Knowing what determines the past, as well as what gives good predictions for the future, gives decision makers better information for making optimal decisions over time In engineering terms, knowing the underlying “laws of motion” of key variables in a dynamic environment... presence of uncertainty” [Wolkenhauer (2001), p xii] This book is about predictive accuracy with neural networks, encompassing forecasting, classification, and dimensionality reduction, and thus involves data engineering.1 The benchmark against which we compare neural network performance is the time-honored linear regression model This model is the starting point of any econometric modeling course, and... fall in reaction to changes in public or private news, because traders are reacting to news and buying or selling assets Approximating the true underlying model means taking into account, as we formulate our models, how traders — human beings like us — actually learn, process information, and make decisions Recent research in macroeconomics by Sargent (1997, 1999), to be discussed in greater detail in. .. but asymmetric and nonlinear, to changes in external variables Neural networks approximate this behavior of economic and financial decision making in a very intuitive way In this important sense neural networks are different from classical econometric models In the neural network model, one is not making any specific hypothesis about the values of the coefficients to be estimated in the model, nor, for... of the cognitive and brain science disciplines for approximating how information is processed and becomes insight We illustrate this point in greater detail when we examine the structure of typical neural network frameworks Suffice it to say, neural network analysis is becoming a key component of the epistemology (philosophy of knowledge) implicit in empirical finance 1.3 The Interface Problems The goal... calculus It is intended for upper-level undergraduate students, beginning graduate students in economics or finance, and professionals working in business and financial research settings The explanation attempts to be straightforward: what these methods are, how they work, and what they can deliver for forecasting and decision making in financial markets The book is not intended for ordinary M.B.A students,... efficiently distill information from these market signals for better diversification and effective hedging, or even better stabilization policy? All of these issues may be addressed very effectively with neural network methods Neural networks help us to approximate or “engineer” data, which, in the words of Wolkenhauer, is both the “art of turning data into information” and “reasoning about data in the presence .