Forecasting the NYSE composite index with technical analysis, pattern recognizer, neural network, and genetic algorithm: a case study in romantic decision support

The 21st century is seeing technological advances that make it possible to build more robust and sophisticated decision support systems than ever before. But the effectiveness of these systems may be limited if we do not consider more eclectic (or romantic) options. This paper exemplifies the potential that lies in the novel application and combination of methods, in this case to evaluating stock market purchasing opportunities using the “technical analysis” school of stock market prediction. Members of the technical analysis school predict market prices and movements based on the dynamics of market price and volume, rather than on economic fundamentals such as earnings and market share. The results of this paper support the effectiveness of the technical analysis approach through use of the “bull flag” price and volume pattern heuristic. The romantic approach to decision support exemplified in this paper is made possible by the recent development of: (1) high-performance desktop computing, (2) the methods and techniques of machine learning and soft computing, including neural networks and genetic algorithms, and (3) approaches recently developed that combine diverse classification and forecasting systems. The contribution of this paper lies in the novel application and combination of the decision-making methods and in the nature and superior quality of the results achieved.     

Forecasting time series using principal component analysis with respect to instrumental variables

Two new forecasting methods of time series are introduced. They are both based on a factorial analysis method called spline principal component analysis with respect to instrumental variables (spline PCAIV). The first method is a straightforward application of spline PCAIV while the second one is an adaptation of spline PCAIV. In the modified version, the used criteria according to the unknown value that need to be predicted are differentiated. Those two forecasting methods are shown to be well adapted to time series.     

Prediction of Road Traffic using a Neural Network Approach

A key component of the daily operation and planning activities of a traffic control centre is short-term forecasting, i.e. the prediction of daily to the next few days of traffic flow. Such forecasts have a significant impact on the optimal regulation of the road traffic on all kinds of freeways. They are increasingly important in an environment with increasing road traffic problems. The present paper aims at presenting the effectiveness of a neural network system for prediction based on time-series data. We only use one parameter, namely traffic volume for the forecasting. We employ artificial neural networks for traffic forecasting applied on a road section. Recurrent Jordan networks, popular in the modelling of time series, is examined in this study. Simulation results demonstrate that learning with this type of architecture has a good generalisation ability.     

Forecasting airborne pollen concentration time series with neural and neuro-fuzzy models

Forecasting airborne pollen concentrations is one of the most studied topics in aerobiology, due to its crucial application to allergology. The most used tools for this problem are single lineal regressions and autoregressive models (ARIMA). Notwithstanding, few works have used more sophisticated tools based in Artificial Intelligence, as are neural or neuro-fuzzy models. In this work, we applied some of these models to forecast olive pollen concentrations in the atmosphere of Granada (Spain). We first studied the overall performance of the selected models, then considering the data segmented into intervals (low, medium and high concentration), to test how they behave on each interval. Experimental results show an advantage of the neuro-fuzzy models against classical statistical methods, although there is still room for improvement.¹     

基于小波分析的预测算法在智能商务中的应用

针对智能商务系统数据挖掘的特性，应用小波分析理论，提出了基于小波分析的预测算法，并将其运用于“天商-2000”智能商务系统中。     

基于ARIMA和神经网络模型的城市燃气负荷预测

城市燃气负荷预测是城市天然气调配的重要环节。在对燃气负荷时间序列进行小波周期分析的基础上,建立燃气负荷的基于ARIMA的神经网络温度矫正模型,ARIMA模型对年周期数据进行平滑,有效去除了过去的短期影响;将大气温度作为神经网络的输入对ARIMA模型预测值进行修正。经过检验,该模型很好地揭示了燃气负荷时间序列的特征,预测效果较好。     

Forecasting method for weak trend quantities

Sometimes the output or input quantities of a system are subjected to random variations incurred from surrounding environments. In such a case, the available historical quantities can not be fitted to one of the known patterns. Therefore, the forecasting about future quantities becomes a stochastic problem. In other words, the methods of time series analysis, in any way, can't be conducted directly to the given data. This paper presents an approximation method for forecasting when the original historical quantities don't fit a trend. The method is mainly based on the assumption that the quantities constitute a frequency distribution with repetitive folds; each of them can be explained by a probability distribution at a specific significance level.     

包含自发自用电量情况的负荷预测

本文针对一种含有自发电的特殊情况来进行负荷预测，对时间序列法在负荷预测的建模中应用，提出了一种ARMAX的模型，结合预测模型，将自发电作为一个大扰动来考虑。通过验证，得到了较好的结果。     

基于熵的自适应门限小波去噪股市价格预测

本文针对传统软阈值法小波去噪采用统一门限而引起的过平滑问题,根据熵的特性,在各层自适应调整去噪门限,提出一种改进的小波去噪算法,采用Hurst指数和盒维数作为判决准则抑制过平滑。最后将算法应用于股市价格时间序列去噪,并用BP神经网络对去噪后的深发展A近20年的收盘价格进行了分段预测。仿真表明,本文方法与传统方法相比,误差明显减小,预测结果更为理想。     

基于周期性建模的时间序列预测方法及电价预测研究

时间序列数据广泛存在于人类的生产生活中, 通常具有复杂的非线性动态和一定的周期性. 与传统的时间序列分析方法相比, 基于深度学习的方法更能捕捉数据的深层特性, 对具有复杂非线性的时间序列有较好的建模效果. 为了在神经网络中显式地建模时间序列数据的周期性和趋势性, 本文在循环神经网络的基础上引入了周期损失和趋势损失, 建立了基于周期性建模和多任务学习的时间序列预测模型. 将模型应用到欧洲能源交易所法国市场的能源市场价格预测中, 结果表明周期损失和趋势损失能够提高神经网络的泛化能力, 并提高预测时间序列趋势的精度.     

Forecasting Financial Time Series using Neural Network and Fuzzy System-based Techniques

Forecasting currency exchange rates are an important financial problem that is receiving increasing attention, especially because of its intrinsic difficulty and practical applications. During the last few years, a number of nonlinear models have been proposed for obtaining accurate prediction results, in an attempt to ameliorate the performance of the traditional linear approaches. Among them, neural network models have been used with encouraging results. This paper presents improved neural network and fuzzy models used for exchange rate prediction. Several approaches, including multi-layer perceptions, radial basis functions, dynamic neural networks and neuro-fuzzy systems, have been proposed and discussed. Their performances for one-step and multiple step ahead predictions have been evaluated through a study, using real exchange daily rate values of the US Dollar vs. British Pound. ID="A1" Correspondence and offprint requests to: Dr V. Kodogiannis, Mechatronics Group, Department of Computer Science, University of Westminster, London HAI 3TP, UK. Email: kodogiv@wmin.ac.uk     

A hybrid modular neural network architecture with fuzzy Sugeno integration for time series forecasting

We describe in this paper the application of a modular neural network architecture to the problem of simulating and predicting the dynamic behavior of complex economic time series. We use several neural network models and training algorithms to compare the results and decide at the end, which one is best for this application. We also compare the simulation results with the traditional approach of using a statistical model. In this case, we use real time series of prices of consumer goods to test our models. Real prices of tomato in the U.S. show complex fluctuations in time and are very complicated to predict with traditional statistical approaches. For this reason, we have chosen a neural network approach to simulate and predict the evolution of these prices in the U.S. market.     

基于多规则实时学习神经网络的时间序列预测模型

在总结现有神经网络方法缺陷的基础上，提出了模型的思路：预测网络小型化；实时学习；多次预测取均值；加入规则辅助神经网络预测。相对于传统的神经网络模型来讲，该模型突出了动态学习、动态预测的特色，增加了辅助预测的3大规则（异常处理规则、再学习规则和取均值规则）。给出了该模型的工作流程，并以一个实际问题说明了该模型训练、预测的全过程。数据实例表明，该模型是正确的、可行的。同时和其他5种模型预测结果的对比表明，该模型的预测结果是最优的，这充分体现了模型的有效性、先进性。     

Use of neural networks for predictions using time series: Illustration with the El Niño Southern oscillation phenomenon

This paper introduces the use of sets of multiple networks (bundled networks) to manage the variability due to different initialization parameters. This method makes it statistically impossible for the networks to be trapped in the same local minimum, and therefore allows better control of the confidence of the prediction eventually given. The spread of the forecasts given by these different networks can be used for prediction reliability purposes. An illustration of this usage is given with the El Niño phenomenon.     

Multi-step Learning Rule for Recurrent Neural Models: An Application to Time Series Forecasting

Multi-step prediction is a difficult task that has attracted increasing interest in recent years. It tries to achieve predictions several steps ahead into the future starting from current information. The interest in this work is the development of nonlinear neural models for the purpose of building multi-step time series prediction schemes. In that context, the most popular neural models are based on the traditional feedforward neural networks. However, this kind of model may present some disadvantages when a long-term prediction problem is formulated because they are trained to predict only the next sampling time. In this paper, a neural model based on a partially recurrent neural network is proposed as a better alternative. For the recurrent model, a learning phase with the purpose of long-term prediction is imposed, which allows to obtain better predictions of time series in the future. In order to validate the performance of the recurrent neural model to predict the dynamic behaviour of the series in the future, three different data time series have been used as study cases. An artificial data time series, the logistic map, and two real time series, sunspots and laser data. Models based on feedforward neural networks have also been used and compared against the proposed model. The results suggest than the recurrent model can help in improving the prediction accuracy.     

多变量时间序列的建模预报及其应用

本文针对推广的Box-Jenkins型多变量传递函数模型,提出了实用的建模预报新算法.算法中包括本文提出的噪声统计特性指数加权有限记忆自适应估计器(EWLMAE)和双置信区间的不良数据检测方法,以及估计预报算式.相对于其他算法,因算法全部采用线性算式,系统部分和噪声部分参数同时估计,故计算较简单,收敛速度较快.仿真和应用结果验证了所提算法的有效性.     

BP神经网络用于藻类生长预测的改进

针对传统的BP神经网络应用于藻类生长预测时,往往出现训练时间较长、输出数据精度低等问题.本文提出了在含有两层隐含层的BP神经网络结构中,对于数量一定的神经元,若神经元在隐含层分配合理,则BP神经网络可以达到减少训练次数并且能满足问题精度的要求.应用实例表明,该方法对预测藻类生长显得非常有效.     

Forecasting the volatility of stock price index

Accurate volatility forecasting is the core task in the risk management in which various portfolios’ pricing, hedging, and option strategies are exercised. Prior studies on stock market have primarily focused on estimation of stock price index by using financial time series models and data mining techniques. This paper proposes hybrid models with neural network and time series models for forecasting the volatility of stock price index in two view points: deviation and direction. It demonstrates the utility of the hybrid model for volatility forecasting. This model demonstrates the utility of the neural network forecasting combined with time series analysis for the financial goods.     

精确在线支持向量回归在股指预测中的应用

建立了基于精确在线支持向量机回归算法的股指短期预测模型,并通过和另外两种基于传统训练方式的支持向量机预测模型进行比较,验证了该方法的有效性。     

Neural network ensemble operators for time series forecasting

The combination of forecasts resulting from an ensemble of neural networks has been shown to outperform the use of a single “best” network model. This is supported by an extensive body of literature, which shows that combining generally leads to improvements in forecasting accuracy and robustness, and that using the mean operator often outperforms more complex methods of combining forecasts. This paper proposes a mode ensemble operator based on kernel density estimation, which unlike the mean operator is insensitive to outliers and deviations from normality, and unlike the median operator does not require symmetric distributions. The three operators are compared empirically and the proposed mode ensemble operator is found to produce the most accurate forecasts, followed by the median, while the mean has relatively poor performance. The findings suggest that the mode operator should be considered as an alternative to the mean and median operators in forecasting applications. Experiments indicate that mode ensembles are useful in automating neural network models across a large number of time series, overcoming issues of uncertainty associated with data sampling, the stochasticity of neural network training, and the distribution of the forecasts.