Modeling and Prediction of the CNY Exchange Rates Using RBF Neural Networks versus GARCH Models

The CNY exchange rates can be viewed as financial time series which are characterized by high uncertainty, nonlinearity and time-varying behavior. Predictions for CNY exchange rates of GBP-CNY and USD-CNY were carried out respectively by means of RBF neural network forecasters and GARCH models. GARCH is a mechanism that includes past variances in the explanation of future variances and a time-series technique that we use to model the serial dependence of volatility. The detailed design of architectures of RBF neural network models, transfer functions of the hidden layer nodes, input vectors and output vectors were made with many tests. While experimental results show that the performance of RBF neural networks for forecasting spot CNY exchange rates is better than that of GARCH, both of them are acceptable and effective especially in short term predictions.

Download Full-text

Multiple Ensemble Neural Network Models with Fuzzy Response Aggregation for Predicting COVID-19 Time Series: The Case of Mexico

Healthcare ◽

10.3390/healthcare8020181 ◽

2020 ◽

Vol 8 (2) ◽

pp. 181 ◽

Cited By ~ 5

Author(s):

Patricia Melin ◽

Julio Cesar Monica ◽

Daniela Sanchez ◽

Oscar Castillo

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Fuzzy Logic ◽

Network Models ◽

Neural Network Models ◽

Validation Data ◽

Data Set ◽

Ensemble Neural Networks ◽

Ensemble Neural Network

In this paper, a multiple ensemble neural network model with fuzzy response aggregation for the COVID-19 time series is presented. Ensemble neural networks are composed of a set of modules, which are used to produce several predictions under different conditions. The modules are simple neural networks. Fuzzy logic is then used to aggregate the responses of several predictor modules, in this way, improving the final prediction by combining the outputs of the modules in an intelligent way. Fuzzy logic handles the uncertainty in the process of making a final decision about the prediction. The complete model was tested for the case of predicting the COVID-19 time series in Mexico, at the level of the states and the whole country. The simulation results of the multiple ensemble neural network models with fuzzy response integration show very good predicted values in the validation data set. In fact, the prediction errors of the multiple ensemble neural networks are significantly lower than using traditional monolithic neural networks, in this way showing the advantages of the proposed approach.

Download Full-text

Evolutionary product unit based neural networks for hydrological time series analysis

Journal of Hydroinformatics ◽

10.2166/hydro.2010.099 ◽

2010 ◽

Vol 13 (4) ◽

pp. 825-841 ◽

Cited By ~ 6

Author(s):

Dulakshi S. K. Karunasingha ◽

A. W. Jayawardena ◽

W. K. Li

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Meteorological Data ◽

Time Series Prediction ◽

Network Models ◽

Engineering Analysis ◽

Neural Network Models ◽

Mlp Neural Network ◽

Evolutionary Product

Artificial Neural Networks (ANNs) are now widely used in many areas of science, medicine, finance and engineering. Analysis and prediction of time series of hydrological/and meteorological data is one such application. Problems that still exist in the application of ANN's are the lack of transparency and the expertise needed for training. An evolutionary algorithm-based method to train a type of neural networks called Product Units Based Neural Networks (PUNN) has been proposed in a 2006 study. This study investigates the applicability of this type of neural networks to hydrological time series prediction. The technique, with a few small changes to improve the performance, is applied to some benchmark time series as well as to a real hydrological time series for prediction. The results show that evolutionary PUNN produce more transparent models compared to widely used multilayer perceptron (MLP) neural network models. It is also seen that training of PUNN models requires less expertise compared to MLPs.

Download Full-text

Dynamic Foreign Currency Trading Guided by Adaptive Forecasting

Review of Pacific Basin Financial Markets and Policies ◽

10.1142/s0219091598000247 ◽

1998 ◽

Vol 01 (03) ◽

pp. 383-418 ◽

Cited By ~ 1

Author(s):

An-Sing Chen ◽

Mark T. Leung

Keyword(s):

Neural Network ◽

Time Series ◽

Exchange Rates ◽

Foreign Currency ◽

Multivariate Time Series ◽

Network Models ◽

Forecast Model ◽

General Regression Neural Network ◽

Neural Network Models ◽

Currency Trading

The difficulty in predicting exchange rates has been a long-standing problem in international finance as most standard econometric methods are unable to produce significantly better forecasts than the random walk model. Recent studies provide some evidence for the ability of multivariate time-series models to generate better forecasts. At the same time, artificial neural network models have been emerging as alternatives to predict exchange rates. In this paper we propose a nonlinear forecast model combining the neural network with the multivariate econometric framework. This hybrid model contains two forecasting stages. A time series approach based on Bayesian Vector Autoregression (BVAR) models is applied to the first stage of forecasting. The estimates from BVAR are then used by the nonparametric General Regression Neural Network (GRNN) to generate enhanced forecasts. To evaluate the economic impact of forecasts, we develop a set of currency trading rules guided by these models. The optimal conditions implied by the investment rules maximize the expected profits given the expected changes in exchange rates and the interest rate differentials between domestic and foreign countries. Both empirical and simulation experiments suggest that the proposed nonlinear adaptive forecasting model not only produces better forecasts but also results in higher investment returns than other types of models. The effect of risk aversion is also considered in the investment simulation.

Download Full-text

Development of models for maximum and time variation of storm surges at the Tanshui estuary

Natural Hazards and Earth System Science ◽

10.5194/nhess-14-2313-2014 ◽

2014 ◽

Vol 14 (9) ◽

pp. 2313-2320 ◽

Cited By ~ 3

Author(s):

C.-P. Tsai ◽

C.-Y. You

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Time Variation ◽

Network Models ◽

Storm Surges ◽

Physical Parameters ◽

Neural Network Models ◽

Modeling And Forecasting ◽

Time Variations

Abstract. In this study, artificial neural networks, including both multilayer perception and the radial basis function neural networks, are applied for modeling and forecasting the maximum and time variation of storm surges at the Tanshui estuary in Taiwan. The physical parameters, including both the local atmospheric pressure and the wind field factors, for finding the maximum storm surges, are first investigated based on the training of neural networks. Then neural network models for forecasting the time series of storm surges are accordingly developed using the major meteorological parameters with time variations. The time series of storm surges for six typhoons were used for training and testing the models, and data for three typhoons were used for model forecasting. The results show that both neural network models perform very well for the forecasting of the time variation of storm surges.

Download Full-text

On the prediction of solar activity using different neural network models

Annales Geophysicae ◽

10.1007/s00585-996-0020-z ◽

1996 ◽

Vol 14 (1) ◽

pp. 20-26 ◽

Cited By ~ 11

Author(s):

F. Fessant ◽

S. Bengio ◽

D. Collobert

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Solar Activity ◽

Time Series Prediction ◽

Network Models ◽

Accurate Prediction ◽

Experimental Results ◽

Neural Network Models ◽

Telecommunication Companies

Abstract. Accurate prediction of ionospheric parameters is crucial for telecommunication companies. These parameters rely strongly on solar activity. In this paper, we analyze the use of neural networks for sunspot time series prediction. Three types of models are tested and experimental results are reported for a particular sunspot time series: the IR5 index.

Download Full-text

EVOLVING RBF NEURAL NETWORKS FOR ADAPTIVE SOFT-SENSOR DESIGN

International Journal of Neural Systems ◽

10.1142/s0129065713500299 ◽

2013 ◽

Vol 23 (06) ◽

pp. 1350029 ◽

Cited By ~ 34

Author(s):

ALEX ALEXANDRIDIS

Keyword(s):

Neural Networks ◽

Fuzzy Inference ◽

Rbf Neural Network ◽

Network Models ◽

Recursive Least Squares ◽

Rbf Neural Networks ◽

Soft Sensors ◽

Neural Network Models ◽

Inference System ◽

Rbf Network

This work presents an adaptive framework for building soft-sensors based on radial basis function (RBF) neural network models. The adaptive fuzzy means algorithm is utilized in order to evolve an RBF network, which approximates the unknown system based on input–output data from it. The methodology gradually builds the RBF network model, based on two separate levels of adaptation: On the first level, the structure of the hidden layer is modified by adding or deleting RBF centers, while on the second level, the synaptic weights are adjusted with the recursive least squares with exponential forgetting algorithm. The proposed approach is tested on two different systems, namely a simulated nonlinear DC Motor and a real industrial reactor. The results show that the produced soft-sensors can be successfully applied to model the two nonlinear systems. A comparison with two different adaptive modeling techniques, namely a dynamic evolving neural-fuzzy inference system (DENFIS) and neural networks trained with online backpropagation, highlights the advantages of the proposed methodology.

Download Full-text

Nonlinear Time Series and Neural-Network Models of Exchange Rates between the US Dollar and Major Currencies

Risks ◽

10.3390/risks4010007 ◽

2016 ◽

Vol 4 (1) ◽

pp. 7 ◽

Cited By ~ 7

Author(s):

David Allen ◽

Michael McAleer ◽

Shelton Peiris ◽

Abhay Singh

Keyword(s):

Neural Network ◽

Time Series ◽

Exchange Rates ◽

Network Models ◽

Nonlinear Time Series ◽

Neural Network Models ◽

The Us ◽

Us Dollar

Download Full-text

NEURAL NETWORKS IN FINANCE AND ECONOMICS FORECASTING

International Journal of Information Technology & Decision Making ◽

10.1142/s021962200700237x ◽

2007 ◽

Vol 06 (01) ◽

pp. 113-140 ◽

Cited By ~ 49

Author(s):

WEI HUANG ◽

KIN KEUNG LAI ◽

YOSHITERU NAKAMORI ◽

SHOUYANG WANG ◽

LEAN YU

Keyword(s):

Neural Network ◽

Economic Growth ◽

Neural Networks ◽

Stock Market ◽

Exchange Rates ◽

Network Models ◽

Stock Market Index ◽

Neural Network Models ◽

Market Index ◽

Comparison Results

Artificial neural networks (ANNs) have been widely applied to finance and economic forecasting as a powerful modeling technique. By reviewing the related literature, we discuss the input variables, type of neural network models, performance comparisons for the prediction of foreign exchange rates, stock market index and economic growth. Economic fundamentals are important in driving exchange rates, stock market index price and economic growth. Most neural network inputs for exchange rate prediction are univariate, while those for stock market index prices and economic growth predictions are multivariate in most cases. There are mixed comparison results of forecasting performance between neural networks and other models. The reasons may be the difference of data, forecasting horizons, types of neural network models and so on. Prediction performance of neural networks can be improved by being integrated with other technologies. Nonlinear combining forecasting by neural networks also provides encouraging results.

Download Full-text

Neural Network Techniques for Time Series Prediction: A Review

JOIV International Journal on Informatics Visualization ◽

10.30630/joiv.3.3.281 ◽

2019 ◽

Vol 3 (3) ◽

Author(s):

Muhammad Faheem Mushtaq ◽

Urooj Akram ◽

Muhammad Aamir ◽

Haseeb Ali ◽

Muhammad Zulqarnain

Keyword(s):

Neural Network ◽

Neural Networks ◽

Time Series ◽

Time Series Data ◽

Weather Prediction ◽

Time Series Prediction ◽

Series Data ◽

Prediction Problem ◽

Neural Network Models ◽

Physical Time

It is important to predict a time series because many problems that are related to prediction such as health prediction problem, climate change prediction problem and weather prediction problem include a time component. To solve the time series prediction problem various techniques have been developed over many years to enhance the accuracy of forecasting. This paper presents a review of the prediction of physical time series applications using the neural network models. Neural Networks (NN) have appeared as an effective tool for forecasting of time series. Moreover, to resolve the problems related to time series data, there is a need of network with single layer trainable weights that is Higher Order Neural Network (HONN) which can perform nonlinearity mapping of input-output. So, the developers are focusing on HONN that has been recently considered to develop the input representation spaces broadly. The HONN model has the ability of functional mapping which determined through some time series problems and it shows the more benefits as compared to conventional Artificial Neural Networks (ANN). The goal of this research is to present the reader awareness about HONN for physical time series prediction, to highlight some benefits and challenges using HONN.

Download Full-text