Forecasting and trading currency volatility: an application of recurrent neural regression and model combination

In this paper, we examine the use of non‐parametric Neural Network Regression (NNR) and Recurrent Neural Network (RNN) regression models for forecasting and trading currency volatility, with an application to the GBP/USD and USD/JPY exchange rates. Both the results of the NNR and RNN models are benchmarked against the simpler GARCH alternative and implied volatility. Two simple model combinations are also analysed. The intuitively appealing idea of developing a nonlinear nonparametric approach to forecast FX volatility, identify mispriced options and subsequently develop a trading strategy based upon this process is implemented for the first time on a comprehensive basis. Using daily data from December 1993 through April 1999, we develop alternative FX volatility forecasting models. These models are then tested out‐of‐sample over the period April 1999–May 2000, not only in terms of forecasting accuracy, but also in terms of trading efficiency: in order to do so, we apply a realistic volatility trading strategy using FX option straddles once mispriced options have been identified. Allowing for transaction costs, most trading strategies retained produce positive returns. RNN models appear as the best single modelling approach yet, somewhat surprisingly, model combination which has the best overall performance in terms of forecasting accuracy, fails to improve the RNN‐based volatility trading results. Another conclusion from our results is that, for the period and currencies considered, the currency option market was inefficient and/or the pricing formulae applied by market participants were inadequate. Copyright © 2002 John Wiley & Sons, Ltd.     

Long‐term sales forecasting using holt–winters and neural network methods

The problem of medium to long‐term sales forecasting raises a number of requirements that must be suitably addressed in the design of the employed forecasting methods. These include long forecasting horizons (up to 52 periods ahead), a high number of quantities to be forecasted, which limits the possibility of human intervention, frequent introduction of new articles (for which no past sales are available for parameter calibration) and withdrawal of running articles. The problem has been tackled by use of a damped‐trend Holt–Winters method as well as feedforward multilayer neural networks (FMNNs) applied to sales data from two German companies. Copyright © 2005 John Wiley & Sons, Ltd.     

Long‐term streamflow forecasting using artificial neural network based on preprocessing technique

Artificial neural network (ANN) combined with signal decomposing methods is effective for long‐term streamflow time series forecasting. ANN is a kind of machine learning method utilized widely for streamflow time series, and which performs well in forecasting nonstationary time series without the need of physical analysis for complex and dynamic hydrological processes. Most studies take multiple factors determining the streamflow as inputs such as rainfall. In this study, a long‐term streamflow forecasting model depending only on the historical streamflow data is proposed. Various preprocessing techniques, including empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD) and discrete wavelet transform (DWT), are first used to decompose the streamflow time series into simple components with different timescale characteristics, and the relation between these components and the original streamflow at the next time step is analyzed by ANN. Hybrid models EMD‐ANN, EEMD‐ANN and DWT‐ANN are developed in this study for long‐term daily streamflow forecasting, and performance measures root mean square error (RMSE), mean absolute percentage error (MAPE) and Nash–Sutcliffe efficiency (NSE) indicate that the proposed EEMD‐ANN method performs better than EMD‐ANN and DWT‐ANN models, especially in high flow forecasting.     

Direction‐of‐Change Financial Time Series Forecasting using a Similarity‐Based Classification Model

A similarity‐based classification model is proposed whereby densities of positive and negative returns in a delay‐embedded input space are estimated from a graphical representation of the data using an eigenvector centrality measure, and subsequently combined under Bayes' theorem to predict the probability of upward/downward movements. Application to directional forecasting of the daily close price of the Dow Jones Industrial Average over a 20‐year out‐of‐sample period yields performance superior to random walk and logistic regression models, and on a par with that of multilayer perceptrons. A feature of the classifier is that it is parameter free, parameters entering the model only via the measure used to determine pairwise similarity between data points. This allows intuitions about the nature of time series to be elegantly integrated into the model. The recursive nature of eigenvector centrality makes it better able to deal with sparsely populated input spaces than conventional approaches based on density estimation. Copyright © 2013 John Wiley & Sons, Ltd.     

日径流预报贝叶斯回声状态网络方法

回声状态网络（ESN）相比传统递归神经网络,具有模型简单、参数训练速度快的特点．针对标准ESN因常采用线性回归率定模型参数容易出现过拟合问题,提出了基于贝叶斯回声状态网络（BESN）的日径流预报模型．该模型将贝叶斯理论与ESN模型相结合,通过权重后验概率密度最大化而获得最优输出权重,提高了模型的泛化能力．通过安砂和新丰江两座水库日径流预测实例表明,BESN模型是一种有效、可行的预测方法,与传统BP神经网络和ESN模型对比,进一步表明BESN模型具有更好的预测精度．     

Mortality forecasting using neural networks and an application to cause‐specific data for insurance purposes

Mortality forecasting is important for life insurance policies, as well as in other areas. Current techniques for forecasting mortality in the USA involve the use of the Lee–Carter model, which is primarily used without regard to cause. A method for forecasting morality is proposed which involves the use of neural networks. A comparative analysis is done between the Lee–Carter model, linear trend and the proposed method. The results confirm that the use of neural networks performs better than the Lee–Carter and linear trend model within 5% error. Furthermore, mortality rates and life expectancy were formulated for individuals with a specific cause based on prevalence data. The rates are broken down further into respective stages (cancer) based on the individual's diagnosis. Therefore, this approach allows life expectancy to be calculated based on an individual's state of health. Copyright © 2008 John Wiley & Sons, Ltd.     

An ensemble of LSTM neural networks for high‐frequency stock market classification

We propose an ensemble of long–short‐term memory (LSTM) neural networks for intraday stock predictions, using a large variety of technical analysis indicators as network inputs. The proposed ensemble operates in an online way, weighting the individual models proportionally to their recent performance, which allows us to deal with possible nonstationarities in an innovative way. The performance of the models is measured by area under the curve of the receiver operating characteristic. We evaluate the predictive power of our model on several US large‐cap stocks and benchmark it against lasso and ridge logistic classifiers. The proposed model is found to perform better than the benchmark models or equally weighted ensembles.     

Non‐linear,non‐parametric,non‐fundamental exchange rate forecasting

This paper employs a non‐parametric method to forecast high‐frequency Canadian/US dollar exchange rate. The introduction of a microstructure variable, order flow, substantially improves the predictive power of both linear and non‐linear models. The non‐linear models outperform random walk and linear models based on a number of recursive out‐of‐sample forecasts. Two main criteria that are applied to evaluate model performance are root mean squared error (RMSE) and the ability to predict the direction of exchange rate moves. The artificial neural network (ANN) model is consistently better in RMSE to random walk and linear models for the various out‐of‐sample set sizes. Moreover, ANN performs better than other models in terms of percentage of correctly predicted exchange rate changes. The empirical results suggest that optimal ANN architecture is superior to random walk and any linear competing model for high‐frequency exchange rate forecasting. Copyright © 2006 John Wiley & Sons, Ltd.     

Estimating the Out‐of‐Sample Predictive Ability of Trading Rules: A Robust Bootstrap Approach

In this paper, we provide a novel way to estimate the out‐of‐sample predictive ability of a trading rule. Usually, this ability is estimated using a sample‐splitting scheme, true out‐of‐sample data being rarely available. We argue that this method makes poor use of the available data and creates data‐mining possibilities. Instead, we introduce an alternative.632 bootstrap approach. This method enables building in‐sample and out‐of‐sample bootstrap datasets that do not overlap but exhibit the same time dependencies. We show in a simulation study that this technique drastically reduces the mean squared error of the estimated predictive ability. We illustrate our methodology on IBM, MSFT and DJIA stock prices, where we compare 11 trading rules specifications. For the considered datasets, two different filter rule specifications have the highest out‐of‐sample mean excess returns. However, all tested rules cannot beat a simple buy‐and‐hold strategy when trading at a daily frequency. Copyright © 2015 John Wiley & Sons, Ltd.     

A neural network versus Black–Scholes: a comparison of pricing and hedging performances

An Erratum has been published for this article in Journal of Forecasting 22(6‐7) 2003, 551 The Black–Scholes formula is a well‐known model for pricing and hedging derivative securities. It relies, however, on several highly questionable assumptions. This paper examines whether a neural network (MLP) can be used to find a call option pricing formula better corresponding to market prices and the properties of the underlying asset than the Black–Scholes formula. The neural network method is applied to the out‐of‐sample pricing and delta‐hedging of daily Swedish stock index call options from 1997 to 1999. The relevance of a hedge‐analysis is stressed further in this paper. As benchmarks, the Black–Scholes model with historical and implied volatility estimates are used. Comparisons reveal that the neural network models outperform the benchmarks both in pricing and hedging performances. A moving block bootstrap is used to test the statistical significance of the results. Although the neural networks are superior, the results are sometimes insignificant at the 5% level. Copyright © 2003 John Wiley & Sons, Ltd.     

Empirical analysis and forecasting of volatility dynamics in high‐frequency returns with time‐varying components

We study intraday return volatility dynamics using a time‐varying components approach, and the method is applied to analyze IBM intraday returns. Empirical evidence indicates that with three additive components—a time‐varying mean of absolute returns and two cosine components with time‐varying amplitudes—together they capture very well the pronounced periodicity and persistence behaviors exhibited in the empirical autocorrelation pattern of IBM returns. We find that the long‐run volatility persistence is driven predominantly by daily level shifts in mean absolute returns. After adjusting for these intradaily components, the filtered returns behave much like a Gaussian noise, suggesting that the three‐components structure is adequately specified. Furthermore, a new volatility measure (TCV) can be constructed from these components. Results from extensive out‐of‐sample rolling forecast experiments suggest that TCV fares well in predicting future volatility against alternative methods, including GARCH model, realized volatility and realized absolute value. Copyright © 2009 John Wiley & Sons, Ltd.     

Forecasting volatility with support vector machine‐based GARCH model

Recently, support vector machine (SVM), a novel artificial neural network (ANN), has been successfully used for financial forecasting. This paper deals with the application of SVM in volatility forecasting under the GARCH framework, the performance of which is compared with simple moving average, standard GARCH, nonlinear EGARCH and traditional ANN‐GARCH models by using two evaluation measures and robust Diebold–Mariano tests. The real data used in this study are daily GBP exchange rates and NYSE composite index. Empirical results from both simulation and real data reveal that, under a recursive forecasting scheme, SVM‐GARCH models significantly outperform the competing models in most situations of one‐period‐ahead volatility forecasting, which confirms the theoretical advantage of SVM. The standard GARCH model also performs well in the case of normality and large sample size, while EGARCH model is good at forecasting volatility under the high skewed distribution. The sensitivity analysis to choose SVM parameters and cross‐validation to determine the stopping point of the recurrent SVM procedure are also examined in this study. Copyright © 2009 John Wiley & Sons, Ltd.     

Hybridizing kernel‐based fuzzy c‐means with hierarchical selective neural network ensemble model for business failure prediction

More and more ensemble models are used to forecast business failure. It is generally known that the performance of an ensemble relies heavily on the diversity between each base classifier. To achieve diversity, this study uses kernel‐based fuzzy c‐means (KFCM) to organize firm samples and designs a hierarchical selective ensemble model for business failure prediction (BFP). First, three KFCM methods—Gaussian KFCM (GFCM), polynomial KFCM (PFCM), and Hyper‐tangent KFCM (HFCM)—are employed to partition the financial data set into three data sets. A neural network (NN) is then adopted as a basis classifier for BFP, and three sets, which are derived from three KFCM methods, are used to build three classifier pools. Next, classifiers are fused by the two‐layer hierarchical selective ensemble method. In the first layer, classifiers are ranked based on their prediction accuracy. The stepwise forward selection method is employed to selectively integrate classifiers according to their accuracy. In the second layer, three selective ensembles in the first layer are integrated again to acquire the final verdict. This study employs financial data from Chinese listed companies to conduct empirical research, and makes a comparative analysis with other ensemble models and all its component models. It is the conclusion that the two‐layer hierarchical selective ensemble is good at forecasting business failure.     

A Neuro‐wavelet Model for the Short‐Term Forecasting of High‐Frequency Time Series of Stock Returns

We propose a wavelet neural network (neuro‐wavelet) model for the short‐term forecast of stock returns from high‐frequency financial data. The proposed hybrid model combines the capability of wavelets and neural networks to capture non‐stationary nonlinear attributes embedded in financial time series. A comparison study was performed on the predictive power of two econometric models and four recurrent neural network topologies. Several statistical measures were applied to the predictions and standard errors to evaluate the performance of all models. A Jordan net that used as input the coefficients resulting from a non‐decimated wavelet‐based multi‐resolution decomposition of an exogenous signal showed a consistent superior forecasting performance. Reasonable forecasting accuracy for the one‐, three‐ and five step‐ahead horizons was achieved by the proposed model. The procedure used to build the neuro‐wavelet model is reusable and can be applied to any high‐frequency financial series to specify the model characteristics associated with that particular series. Copyright © 2013 John Wiley & Sons, Ltd.     

Choosing among competing econometric forecasts: Regression‐based forecast combination using model selection

Forecast combination based on a model selection approach is discussed and evaluated. In addition, a combination approach based on ex ante predictive ability is outlined. The model selection approach which we examine is based on the use of Schwarz (SIC) or the Akaike (AIC) Information Criteria. Monte Carlo experiments based on combination forecasts constructed using possibly (misspecified) models suggest that the SIC offers a potentially useful combination approach, and that further investigation is warranted. For example, combination forecasts from a simple averaging approach MSE‐dominate SIC combination forecasts less than 25% of the time in most cases, while other ‘standard’ combination approaches fare even worse. Alternative combination approaches are also compared by conducting forecasting experiments using nine US macroeconomic variables. In particular, artificial neural networks (ANN), linear models, and professional forecasts are used to form real‐time forecasts of the variables, and it is shown via a series of experiments that SIC, t‐statistic, and averaging combination approaches dominate various other combination approaches. An additional finding is that while ANN models may not MSE‐dominate simpler linear models, combinations of forecasts from these two models outperform either individual forecast, for a subset of the economic variables examined. Copyright © 2001 John Wiley & Sons, Ltd.     

Forecasting nonlinear time series with feed‐forward neural networks: a case study of Canadian lynx data

The forecasting capabilities of feed‐forward neural network (FFNN) models are compared to those of other competing time series models by carrying out forecasting experiments. As demonstrated by the detailed forecasting results for the Canadian lynx data set, FFNN models perform very well, especially when the series contains nonlinear and non‐Gaussian characteristics. To compare the forecasting accuracy of a FFNN model with an alternative model, Pitman's test is employed to ascertain if one model forecasts significantly better than another when generating one‐step‐ahead forecasts. Moreover, the residual‐fit spread plot is utilized in a novel fashion in this paper to compare visually out‐of‐sample forecasts of two alternative forecasting models. Finally, forecasting findings on the lynx data are used to explain under what conditions one would expect FFNN models to furnish reliable and accurate forecasts. Copyright © 2005 John Wiley & Sons, Ltd.     

A generalized regression model based on hybrid empirical mode decomposition and support vector regression with back-propagation neural network for mid-short-term load forecasting

Since load forecasting plays a decisive role in the safe and stable operation of power systems, it is particularly important to explore forecasting methods accurately. In this article, the hybrid empirical mode decomposition (EMD) and support vector regression (SVR) with back-propagation neural network (BPNN), namely the EMDHR-SVR-BPNN model, is proposed. Information theory is mainly used to solve the data tendency problem, and the EMD method is used to solve the data volatility problem. There is no interaction between these two methods; thus these two models can complement each other through generalized regression of orthogonal decomposition. Taking the load data from the New South Wales (NSW, Australia) market as an example, the obtained simulation results are compared with other models. It is concluded that the proposed EMDHR-SVR-BPNN model not only improves the forecasting accuracy but also has good fitting ability. It can reflect the changing tendency of data in a timely manner, providing a strong basis for the electricity generation of the power sector in the future, thus reducing electricity waste. The proposed EMDHR-SVR-BPNN model has potential for employment in mid-short term load forecasting.     

Short‐term forecasting of GDP using large datasets: a pseudo real‐time forecast evaluation exercise

This paper performs a large‐scale forecast evaluation exercise to assess the performance of different models for the short‐term forecasting of GDP, resorting to large datasets from ten European countries. Several versions of factor models are considered and cross‐country evidence is provided. The forecasting exercise is performed in a simulated real‐time context, which takes account of publication lags in the individual series. In general, we find that factor models perform best and models that exploit monthly information outperform models that use purely quarterly data. However, the improvement over the simpler, quarterly models remains contained. Copyright © 2009 John Wiley & Sons, Ltd.     

Forecasting economic indicators using a consumer sentiment index: Survey‐based versus text‐based data

Given the confirmed effectiveness of the survey‐based consumer sentiment index (CSI) as a leading indicator of real economic conditions, the CSI is actively used in making policy judgments and decisions in many countries. However, although the CSI offers qualitative information for presenting current conditions and predicting a household's future economic activity, the survey‐based method has several limitations. In this context, we extracted sentiment information from online economic news articles and demonstrated that the Korean cases are a good illustration of applying a text mining technique when generating a CSI using sentiment analysis. By applying a simple sentiment analysis based on the lexicon approach, this paper confirmed that news articles can be an effective source for generating an economic indicator in Korea. Even though cross‐national comparative research results are suited better than national‐level data to generalize and verify the method used in this study, international comparisons are quite challenging to draw due to the necessary linguistic preprocessing. We hope to encourage further cross‐national comparative research to apply the approach proposed in this study.