Time series forecasting using fuzzy transformation and neural network with back propagation learning

Pal, Shanoli Samui; Kar, Samarjit

doi:10.3233/jifs-161767

Cited by 13 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The presented method has been applied in the Composite Index of Shanghai Capital Exchange data gathered for the term January 1992 to December 2008. The conclusion has been compared with the outcome computed by employing the de-noising ability of WT together with BPNN (Pal & Kar, 2017). A multistep-forward prediction approach has been used which mixes empirical mode decomposition (EMD) and SVR.…”

Section: Literature Reviewmentioning

confidence: 99%

“…It can be used for a continuous function on a fixed distance, [a, b] ∈ R the transformation of the fuzzy space of functions on an open space of real numbers by a direct transformation into the real n-dimensional vectors. Their inverse conversion returns the derived n-vector to the initial or approximate function of it (Pal & Kar, 2017). Let x 1 < x 2 < • • • < x n be fixed nodes in [a, b], such that x 1 a, .…”

Section: Fuzzy Transformmentioning

confidence: 99%

See 1 more Smart Citation

Integrating Wavelet Decomposition and Fuzzy Transformation for Improving the Accuracy of Forecasting Crude Oil Price

Saghi

Rezaee

2021

Comput Econ

View full text Add to dashboard Cite

In this paper, hybrid methods are proposed to predict OPEC crude oil. In the preprocessing step, the wavelet decomposition has been used to reduce the noise of time series, which divides the original data into five levels. Also, the fuzzy transform (F-transform) is applied for its potency of management uncertainty due to the fluctuation of data. F-transform is for decomposing the time series data and afterward, this decomposed data are employed on the ground of inputs. Because of the nonlinearity of the time series structure, integrating the mentioned proposed algorithms with the multilayer perceptron (MLP), radial basis functions, group method of data handling (GMDH), adaptive neuro-fuzzy inference system (ANFIS), and support vector regression (SVR) are applied for prediction of time series relevant to the crude oil price. By comparing the results of models, L5-ANFIS, L5-MLP, L5-GMDH, L5-SVR are more appropriate than others for predicting OPEC crude oil price.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Fuzzy Transformmentioning

confidence: 99%

Integrating Wavelet Decomposition and Fuzzy Transformation for Improving the Accuracy of Forecasting Crude Oil Price

Saghi

Rezaee

2021

Comput Econ

View full text Add to dashboard Cite

show abstract

“…There are some recent works on stock market time series data. Deep learning technique has been used in predicting digital currencies, like Bitcoin, Digital Cash and Ripple [24], an ensemble of neural networks (NN) coupled with particle swarm intelligence for parameter optimization has been used in the technical analysis information fusion [22], singular spectrum analysis (SSA) and support vector regression (SVR) coupled with particle swarm optimization (PSO) has been implemented for intraday stock price prediction [23], neural network weight adjustment using zSlices-based generalized type-2 fuzzy set has been applied in predicting closing price index of Shenzhen stock exchange, closing price index of Shanghai stock exchange [32], fuzzy transformation and neural network with back propagation learning has been used in stock market closing price index [30], data discretization using fuzzy statistics and rule generation by rough set theory has been used in stock market time series forecasting [33], an improved fuzzy time series model for unequal interval length using genetic algorithm has been applied on BSE sensex time series and Shenzhen stock exchange data [31].…”

Section: Introductionmentioning

confidence: 99%

Forecasting stock market price by using fuzzified Choquet integral based fuzzy measures with genetic algorithm for parameter optimization

Pal

Kar

2020

RAIRO-Oper. Res.

Self Cite

View full text Add to dashboard Cite

In this paper, fuzzified Choquet integral and fuzzy-valued integrand with respect to separate measures like fuzzy measure, signed fuzzy measure and intuitionistic fuzzy measure are used to develop regression model for forecasting. Fuzzified Choquet integral is used to build a regression model for forecasting time series with multiple attributes as predictor attributes. Linear regression based forecasting models are suffering from low accuracy and unable to approximate the non-linearity in time series. Whereas Choquet integral can be used as a general non-linear regression model with respect to non classical measures. In the Choquet integral based regression model parameters are optimized by using a real coded genetic algorithm (GA). In these forecasting models, fuzzified integrands denote the participation of an individual attribute or a group of attributes to predict the current situation. Here, more generalized Choquet integral, i.e., fuzzified Choquet integral is used in case of non-linear time series forecasting models. Three different real stock exchange data are used to predict the time series forecasting model. It is observed that the accuracy of prediction models highly depends on the non-linearity of the time series.

show abstract

“…With the introduction of the autoregressive integrated moving average (ARIMA) model, the application of regression analysis method in time series forecasting has become mature; however, regression analysis method can be applied only to linear time series (Pannakkong et al 2018). Due to the non-linear characteristics of time series in practice, many scholars have carried out the research on non-linear time series prediction using the back propagation (BP) algorithm (Pal and Kar 2017), support vector machine (SVM) (Xiao et al 2019), radial basis function (RBF) (Awad and Qasrawi 2018), echo state networks (ESN) (Lopez et al 2018;Liang et al 2018), and other models (Yeh et al 2019). The traditional artificial neural networks cannot capture a long-term dependence of time series data.…”

Section: Introductionmentioning

confidence: 99%

A dilated convolution network-based LSTM model for multi-step prediction of chaotic time series

et al. 2019

View full text Add to dashboard Cite

Aiming to solve the problems of low accuracy of multi-step prediction and difficulty in determining the maximum number of prediction steps of chaotic time series, a multi-step time series prediction model based on the dilated convolution network and long short-term memory (LSTM), named the dilated convolution-long short-term memory (DC-LSTM), is proposed. The dilated convolution operation is used to extract the correlation between the predicted variable and correlational variables. The features extracted by dilated convolution operation and historical data of predicted variable are input into LSTM to obtain the desired multi-step prediction result. Furthermore, cross-correlation analyses (CCA) are applied to calculate the reasonable maximum prediction steps of chaotic time series. Actual applications of multi-step prediction were studied to demonstrate the effectiveness of the proposed model which has superiorities in RMSE, MAE and prediction accuracy because of the extraction of correlation between the predicted variable and correlational variables. Moreover, the proposed DC-LSTM model provides a new method for prediction of chaotic time series and lays a foundation for scientific data analysis of chaotic time series monitoring systems.

show abstract

Time series forecasting using fuzzy transformation and neural network with back propagation learning

Cited by 13 publications

References 14 publications

Integrating Wavelet Decomposition and Fuzzy Transformation for Improving the Accuracy of Forecasting Crude Oil Price

Integrating Wavelet Decomposition and Fuzzy Transformation for Improving the Accuracy of Forecasting Crude Oil Price

Forecasting stock market price by using fuzzified Choquet integral based fuzzy measures with genetic algorithm for parameter optimization

A dilated convolution network-based LSTM model for multi-step prediction of chaotic time series

Contact Info

Product

Resources

About