Neural networks have been shown to be a promising tool for forecasting financial time series. Several design factors significantly impact the accuracy of neural network forecasts. These factors include selection of input variables, architecture of the network, and quantity of training data. The questions of input variable selection and system architecture design have been widely researched, but the corresponding question of how much information to use in producing high-quality neural network models has not been adequately addressed. In this paper, the effects of different sizes of training sample sets on forecasting currency exchange rates are examined. It is shown that those neural networks---given an appropriate amount of historical knowledge ---can forecast future currency exchange rates with 60 percent accuracy, while those neural networks trained on a larger training set have a worse forecasting performance. In addition to higher-quality forecasts, the reduced training set sizes reduce development cost and time.

Billions of dollars flow through the world's financial markets every day, and market participants are understandably eager to accurately price financial instruments and understand relationships involving them. Nonlinear multivariate statistical modeling on fast computers offers the potential to capture more of the underlying dynamics of these high dimensional, noisy systems than traditional models while at the same time making fewer restrictive assumptions about them. For this style of exploratory, nonparametric modeling to be useful, however, care must be taken in fundamental estimation and confidence issues, especially concerns deriving from limited sample sizes. This thesis presents a collection of practical techniques to address these issues for a modeling methodology, Radial Basis Function networks. These techniques include efficient methods for parameter estimation and pruning, including a heuristic for setting good initial parameter values, a pointwise prediction error estimator for kernel type RBF networks, and a methodology for controlling the "data mining" problem. Novel applications in the finance area are described, including the derivation of customized, adaptive option pricing formulas that can distill information about the associated time varying systems that may not be readily captured by theoretical models. A second application area is stock price prediction, where models are found with lower out-of-sample error and better "paper trading" profitability than that of simpler linear and/or univariate models, although their true economic significance for real life trading is questionable. Finally, a case is made for fast computer implementations of these ideas to facilitate the necessary model searching and confidence testing, and related implementation iss...

Forecasting exchange rates is an important financial problem that is receiving increasing attention especially because of its difficulty and practical applications. Artificial neural networks (ANNs) have been widely used as a promising alternative approach for a forecasting task because of several distinguished features. Research efforts on ANNs for forecasting exchange rates are considerable. In this paper, we attempt to provide a survey of research in this area. Several design factors significantly impact the accuracy of neural network forecasts. These factors include the selection of input variables, preparing data, and network architecture. There is no consensus about the factors. In different cases, various decisions have their own effectiveness. We also describe the integration of ANNs with other methods and report the comparison between performances of ANNs and those of other forecasting methods, and finding mixed results. Finally, the future research directions in this area are discussed.

We apply neural network ensembles to the task of forecasting financial time series and explore the use of high order statistical information as part of network inputs. We show that the prediction accuracy of the time series can be significanlty improved utilizing this methodology. Since prediction accuracy is only an estimate for the profitability on the financial market, we report good and profitable results using a profit/loss metric based on market simulations. Our simulations show an improvement of between 1.3 to 12.4% over a simple buy and hold trading strategy, and an improvement of between 6.5 to 20.9% over trading strategy using linear autoregressive models. Keywords: Time Series Analysis, Ensemble Networks, Back-Propagation, High Order Statistics 1. Introduction Forecasting financial time series relies on the discovery of strong empirical regularities in observations of the system and has been widely discussed [2][4][5]. Because these regularities are often masked by noise ...

ance-based algorithms to compute distances that may not reflect the optimal distance between two data points. For example, two input features may be identical. The effect of these two identical input features is equivalent to a single feature with twice the weight during distance calculations. The feature's larger weight is only justified if it contains more information with respect to the desired outputs than the other features. Otherwise the larger weight will result in a degradation in classification accuracy. De-correlation of input features may therefore improve the classification accuracy of distance-based 77 Table 4.8. The performance of the weighted vote kNN algorithm without feature weights (kNNwv ), with computed feature weights (kNNwv FWMI ), or learned feature weights (kNNwv FW V SM ). Domain kNNwv kNNwv FWMI kNNwv

There is growing interest both in the field of neural computing and in the financial world in the possibility of using neural networks to forecast the future changes in prices of stocks, exchange rates and commodities. Since networks havebeenshown to be capable of modelling the underlying structure of a time series, many attempts have been made at exploiting that capability in order to carry out atechnical analysis of such prices. If the efficientmarkets hypothesis is true however, there is no underlying structure to be modelled and the whole endevour is doomed to failure. This paper investigates the common methods for such an approach and outlines the major pitfalls and common errors to avoid. It is the author's hope that in pointing out the possible pitfalls now, we can avoid making claims to the commercial world before we are properly ready to do so. 1 Introduction There is a justifiable scepticism surrounding the idea that it is possible to make money by predicting price ...

Forecasting exchange rates is an important financial problem that is receiving increasing attention especially because of its di#- culty and practical applications. This paper proposes a Flexible Neural Tree (FNT) model for forecasting three major international currency exchange rates. Based on the pre-defined instruction/operator sets, a flexible neural tree model can be created and evolved. This framework allows input variables selection, over-layer connections and di#erent activation functions for the various nodes involved. The FNT structure is developed using the Extended Compact Genetic Programming and the free parameters embedded in the neural tree are optimized by particle swarm optimization algorithm. Empirical results indicate that the proposed method is better than the conventional neural network forecasting models.

(RLSR)is a technique originally from Statistical Learning (SL) theory. RLSR can deal with non-linear problem through mapping the samples into a higher dimension space using a kernel function. This paper adopts the RLSR to time series forecasting and the resulted model is termed RLS-TS model getting the idea from applying neural network and support vector regression to time series forecasting. This paper applies the RLS-TS model to GBP/USD Exchange Rate forecasting. RLS-TS performs better than random walk, linear regression, autoregression integrated moving average, and artificial neural network model in predicting GBP/USD currency exchange rates. A grid search is used to choose the optimal parameters. Index Terms—exchange rate, regularized least-squares, time series, forecasting.

Abstract. Time-series prediction can be interpreted in a way that is suitable for artificial intelligence

Abstract not found