A new model for time-series forecasting using radial basis functions and exogenous data

In this paper, we present a new model for time-series forecasting using radial basis functions (RBFs) as a unit of artificial neural networks (ANNs), which allows the inclusion of exogenous information (EI) without additional pre-processing. We begin by summarizing the most well-known EI techniques used ad hoc, i.e., principal component analysis (PCA) and independent component analysis (ICA). We analyze the advantages and disadvantages of these techniques in time-series forecasting using Spanish bank and company stocks. Then, we describe a new hybrid model for time-series forecasting which combines ANNs with genetic algorithms (GAs). We also describe the possibilities when implementing the model on parallel processing systems.

---

A distance-based fuzzy time series model for exchange rates forecasting

Fuzzy time series model has been successfully employed in predicting stock prices and foreign exchange rates. In this paper, we propose a new fuzzy time series model termed as distance-based fuzzy time series (DBFTS) to predict the exchange rate. Unlike the existing fuzzy time series models which require exact match of the fuzzy logic relationships (FLRs), the distance-based fuzzy time series model uses the distance between two FLRs in selecting prediction rules. To predict the exchange rate, a two factors distance-based fuzzy time series model is constructed. The first factor of the model is the exchange rate itself and the second factor comprises many candidate variables affecting the fluctuation of exchange rates. Using the exchange rate data released by the Central Bank of Taiwan, we conducted several experiments on exchange rate forecasting. The experiment results showed that the distance-based fuzzy time series outperformed the random walk model and the artificial neural network model in terms of mean square error.     

A two stage fuzzy piecewise logistic model for penetration forecasting

It is undeniably crucial for a firm to be able to make a forecast regarding the sales volume of new products. However, the current economic environments invariably have uncertain factors and rapid fluctuations where decision makers must draw conclusions from minimal data. Previous studies combine scenario analysis and technology substitution models to forecast the market share of multigenerational technologies. However, a technology substitution model based on a logistic curve will not always fit the S curve well. Therefore, based on historical data and the data forecast by both the Scenario and Delphi methods, a two stage fuzzy piecewise logistic growth model with multiple objective programming is proposed herein. The piecewise concept is adopted in order to reflect the market impact of a new product such that it can be possible to determine the effective length of sales forecasting intervals even when handling a large variation in data or small size data. In order to demonstrate the model's performance, two cases in the Television and Telecommunication industries are treated using the proposed method and the technology substitution model or the Norton and Bass diffusion model. A comparison of the results shows that the proposed model outperforms the technology substitution model and the Norton and Bass diffusion model.     

Applications of type-2 fuzzy logic systems to forecasting of time-series

In this paper, we begin with a type-1 fuzzy logic system (FLS), trained with noisy data. We then demonstrate how information about the noise in the training data can be incorporated into a type-2 FLS, which can be used to obtain bounds within which the true (noisefree) output is likely to lie. We do this with the example of a one-step predictor for the Mackey–Glass chaotic time-series [M.C. Mackey, L. Glass, Oscillation and chaos in physiological control systems, Science 197 (1977) 287–280]. We also demonstrate how a type-2 FLS can be used to obtain better predictions than those obtained with a type-1 FLS.     

一种新的时间序列分析算法及其在股票预测中的应用

分析了股票市场高度非线性的特点,给出了一种改进的时间序列分析算法。新算法利用径向基网络来对序列中的历史信息进行非线性组合,从而比基于线性组合的时间序列分析算法的基本模型更能有效地挖掘出序列中历史信息之间的相互作用。新算法还利用改进的遗传算法对径向基函数的中心和宽度进行了全局范围的优化选择,进一步提高了径向基网络的非线性映射能力。运用该算法对股票走势进行了预测,取得了令人满意的效果。     

基于AR_SVR模型的时间序列预测算法的研究

掌握农产品未来价格变化趋势,有利于正确引导农业生产,提出一种基于自回归与支持向量回归(auto regressive and support vector regression,AR_SVR)模型的非平稳时间序列预测方法.首先,利用AR模型对非平稳时间序列进行季节差分和差分,使其具有平稳性,然后给平稳序列定阶,最后用SVR模型拟合平稳序列,回推得出原始序列的预测值.实验结果表明,AR_SVR模型预测值与真实值很接近,具有较好的预测效果.     

Deterministic fuzzy time series model for forecasting enrollments

The fuzzy time series has recently received increasing attention because of its capability of dealing with vague and incomplete data. There have been a variety of models developed to either improve forecasting accuracy or reduce computation overhead. However, the issues of controlling uncertainty in forecasting, effectively partitioning intervals, and consistently achieving forecasting accuracy with different interval lengths have been rarely investigated. This paper proposes a novel deterministic forecasting model to manage these crucial issues. In addition, an important parameter, the maximum length of subsequence in a fuzzy time series resulting in a certain state, is deterministically quantified. Experimental results using the University of Alabama’s enrollment data demonstrate that the proposed forecasting model outperforms the existing models in terms of accuracy, robustness, and reliability. Moreover, the forecasting model adheres to the consistency principle that a shorter interval length leads to more accurate results.     

A combination of hidden Markov model and fuzzy model for stock market forecasting

This paper presents a novel combination of the hidden Markov model (HMM) and the fuzzy models for forecasting stock market data. In a previous study we used an HMM to identify similar data patterns from the historical data and then used a weighted average to generate a ‘one-day-ahead’ forecast. This paper uses a similar approach to identify data patterns by using the HMM and then uses fuzzy logic to obtain a forecast value. The HMM's log-likelihood for each of the input data vectors is used to partition the dataspace. Each of the divided dataspaces is then used to generate a fuzzy rule. The fuzzy model developed from this approach is tested on stock market data drawn from different sectors. Experimental results clearly show an improved forecasting accuracy compared to other forecasting models such as, ARIMA, artificial neural network (ANN) and another HMM-based forecasting model.     

Dimensionality reduction for multivariate time-series data mining

A multivariate time series is one of the most important objects of research in data mining. Time and variables are two of its distinctive characteristics that add the complication of the algorithms applied to data mining. Reduction in the dimensionality is often regarded as an effective way to address these issues. In this paper, we propose a method based on principal component analysis (PCA) to effectively reduce the dimensionality. We call it “piecewise representation based on PCA” (PPCA), which segments multivariate time series into several sequences, calculates the covariance matrix for each of them in terms of the variables, and employs PCA to obtain the principal components in an average covariance matrix. The results of the experiments, including retained information analysis, classification, and a comparison of the central processing unit time consumption, demonstrate that the PPCA method used to reduce the dimensionality in multivariate time series is superior to the prior methods.

     

A heuristic error-feedback learning algorithm for fuzzy modeling

Describes a type of fuzzy system with interpolating capability to extract MISO fuzzy rules from input-output sample data through learning. The proposed model inherits many merits from Sugeno-type models and their variations. A heuristic error-feedback learning algorithm associated with the model is suggested. Based on which, the estimator is shown to have a self-adjusting step when approaching a minimum     

Fuzzy dual-factor time-series for stock index forecasting

There is an old Wall Street adage goes, “It takes volume to make price move”. The contemporaneous relation between trading volume and stock returns has been studied since stock markets were first opened. Recent researchers such as Wang and Chin [Wang, C. Y., & Chin S. T. (2004). Profitability of return and volume-based investment strategies in China’s stock market. Pacific-Basin Finace Journal, 12, 541–564], Hodgson et al. [Hodgson, A., Masih, A. M. M., & Masih, R. (2006). Futures trading volume as a determinant of prices in different momentum phases. International Review of Financial Analysis, 15, 68–85], and Ting [Ting, J. J. L. (2003). Causalities of the Taiwan stock market. Physica A, 324, 285–295] have found the correlation between stock volume and price in stock markets. To verify this saying, in this paper, we propose a dual-factor modified fuzzy time-series model, which take stock index and trading volume as forecasting factors to predict stock index. In empirical analysis, we employ the TAIEX (Taiwan stock exchange capitalization weighted stock index) and NASDAQ (National Association of Securities Dealers Automated Quotations) as experimental datasets and two multiple-factor models, Chen’s [Chen, S. M. (2000). Temperature prediction using fuzzy time-series. IEEE Transactions on Cybernetics, 30 (2), 263–275] and Huarng and Yu’s [Huarng, K. H., & Yu, H. K. (2005). A type 2 fuzzy time-series model for stock index forecasting. Physica A, 353, 445–462], as comparison models. The experimental results indicate that the proposed model outperforms the listing models and the employed factors, stock index and the volume technical indicator, VR(t), are effective in stock index forecasting.     

A comparison study between fuzzy time series model and ARIMA model for forecasting Taiwan export

This study compares the application of two forecasting methods on the amount of Taiwan export, the ARIMA time series method and the fuzzy time series method. Models discussed for the fuzzy time series method include the Factor models, the Heuristic models, and the Markov model. When the sample period is prolong in our models, the ARIMA model shows smaller than predicted error and closer predicted trajectory to the realistic trend than those of the fuzzy model, resulted in more accurate forecasts of the export amount in the ARIMA model. Especially, the coefficient of the error term for the previous period has increased to 79%, implying the influential effect of external factors. These external factors attribute to the export amount of Taiwan according to the economic viewpoints. However, this impact reduces as time progressing and the export amount of the lag period of 12 or 13 do not affect current export amount anymore. In conclusion, when the sample period is shorter with only a small set of data available, the fuzzy time series models can be utilized to predict export values accurately, outperforming the ARIMA model.     

A multivariate model of fuzzy integrated logical forecasting method (M-FILF) and multiplicative time series clustering: A model of time-varying volatility for dry cargo freight market

The aim of this paper is to improve the fuzzy logical forecasting model (FILF) by utilizing multivariate inference and the partitioning problem for an exponentially distributed time series by using a multiplicative clustering approach. Fuzzy time series (FTS) is a growing study field in computer science and its superiority is indicated frequently. Since the conventional time series analysis requires various pre-conditions, the FTS framework is very useful and convenient for many problems in business practice. This paper particularly investigates pricing problems in the shipping business and price-volatility relationship is the theoretical point of the proposed approach. Both FTS and conventional time series results are comparatively presented in the final section and superiority of the proposed method is explicitly noted.     

A multivariate fuzzy c-means method

Fuzzy c-means (FCMs) is an important and popular unsupervised partitioning algorithm used in several application domains such as pattern recognition, machine learning and data mining. Although the FCM has shown good performance in detecting clusters, the membership values for each individual computed to each of the clusters cannot indicate how well the individuals are classified. In this paper, a new approach to handle the memberships based on the inherent information in each feature is presented. The algorithm produces a membership matrix for each individual, the membership values are between zero and one and measure the similarity of this individual to the center of each cluster according to each feature. These values can change at each iteration of the algorithm and they are different from one feature to another and from one cluster to another in order to increase the performance of the fuzzy c-means clustering algorithm. To obtain a fuzzy partition by class of the input data set, a way to compute the class membership values is also proposed in this work. Experiments with synthetic and real data sets show that the proposed approach produces good quality of clustering.     

Trend time-series modeling and forecasting with neural networks.

Despite its great importance, there has been no general consensus on how to model the trends in time-series data. Compared to traditional approaches, neural networks (NNs) have shown some promise in time-series forecasting. This paper investigates how to best model trend time series using NNs. Four different strategies (raw data, raw data with time index, detrending, and differencing) are used to model various trend patterns (linear, nonlinear, deterministic, stochastic, and breaking trend). We find that with NNs differencing often gives meritorious results regardless of the underlying data generating processes (DGPs). This finding is also confirmed by the real gross national product (GNP) series.     

A new boosting algorithm for improved time-series forecasting with recurrent neural networks

Ensemble methods for classification and regression have focused a great deal of attention in recent years. They have shown, both theoretically and empirically, that they are able to perform substantially better than single models in a wide range of tasks. We have adapted an ensemble method to the problem of predicting future values of time series using recurrent neural networks (RNNs) as base learners. The improvement is made by combining a large number of RNNs, each of which is generated by training on a different set of examples. This algorithm is based on the boosting algorithm where difficult points of the time series are concentrated on during the learning process however, unlike the original algorithm, we introduce a new parameter for tuning the boosting influence on available examples. We test our boosting algorithm for RNNs on single-step-ahead and multi-step-ahead prediction problems. The results are then compared to other regression methods, including those of different local approaches. The overall results obtained through our ensemble method are more accurate than those obtained through the standard method, backpropagation through time, on these datasets and perform significantly better even when long-range dependencies play an important role.     

Applying fuzzy grey modification model on inflow forecasting

This paper investigates a modified grey model for forecasting the inflow of a reservoir. The integral form of the background value is employed for the original grey model, GM(1,1), to improve accuracy and applicability. Thereafter, the Fourier series is altered to handle extreme values with regard to prediction; exponential smoothing is used to improve the drawbacks of the prediction delay phenomenon. Finally, we are hybridised as the ultimate grey model with outstanding prediction accuracy, namely EFGM(1,1). As a typhoon causes significant changes in the inflow of a reservoir, this paper applies the fuzzy membership function for dealing with it during the flood season to construct the fuzzy grey modification model, FEFGM(1,1). Results of grey models are compared with those of the Autoregressive Integrated Moving Average (ARIMA). By evaluating different indices, the errors of the predicted extreme value of EFGM(1,1) perform better than those of GM(1,1) and ARIMA, however worse than that of FEFGM(1,1). The final FEFGM(1,1) shows high precision with regard to reservoir inflow prediction during typhoons with combined effects of fuzzy, exponential smoothing, Fourier series.     

Predictive modular fuzzy systems for time-series classification

We introduce the so-called predictive modular fuzzy system (PREMOFS) which performs time-series classification. A PREMOFS consists of 1) a bank of prediction modules and 2) a fuzzy decision module. It is assumed that the time series is generated by a source belonging to a finite search set (universal set); then the classification problem is to select the source that best represents the observed data, Classification is based on a membership function which is updated recursively according to the predictive accuracy of each model. Two algorithms are presented for updating the membership function. The first is based on sum/product fuzzy inference and the second on max/min fuzzy inference. In short, PREMOFS is a fuzzy modular system that classifies time series to one of a finite number of classes using the full set of past data (without preprocessing) to perform a recursive competitive computation of membership function based on predictive accuracy. Convergence proofs are given for both PREMOFS algorithms; in both cases the membership grade tends to one for the source that best predicts the observed data and to less than one for the remaining sources; hence, correct classification is guaranteed. Simulation results are also presented: PREMOFS are applied to signal detection, system identification, and phoneme classification tasks     

A local linear radial basis function neural network for financial time-series forecasting

In this paper a Local Linear Radial Basis Function Neural Network (LLRBFN) is presented. The difference between the proposed neural network and the conventional Radial Basis Function Neural Network (RBFN) is connection weights between the hidden layer and the output layer which are replaced by a local linear model in the LLRBFN. A modified Particle Swarm Optimization (PSO) with hunter particles is introduced for training the LLRBFN. The proposed methods have been applied for prediction of financial time-series and the result shows the feasibility and effectiveness.     

A modified genetic algorithm for forecasting fuzzy time series

Fuzzy time series approaches are used when observations of time series contain uncertainty. Moreover, these approaches do not require the assumptions needed for traditional time series approaches. Generally, fuzzy time series methods consist of three stages, namely, fuzzification, determination of fuzzy relations, and defuzzification. Artificial intelligence algorithms are frequently used in these stages with genetic algorithms being the most popular of these algorithms owing to their rich operators and good performance. However, the mutation operator of a GA may cause some negative results in the solution set. Thus, we propose a modified genetic algorithm to find optimal interval lengths and control the effects of the mutation operator. The results of applying our new approach to real datasets show superior forecasting performance when compared with those obtained by other techniques.