HYBRID FUZZY NEURAL NETWORK TO PREDICT PRICE DIRECTION IN THE GERMAN DAX-30 INDEX

Intraday trading rules require accurate information about the future short term market evolution. For that reason, next-day market trend prediction has attracted the attention of both academics and practitioners. This interest has increased in recent years, as different methodologies have been applied to this end. Usually, machine learning techniques are used such as artificial neural networks, support vector machines and decision trees. The input variables of most of the studies are traditional technical indicators which are used by professional traders to implement investment strategies. We analyse if these indicators have predictive power on the German DAX-30 stock index by applying a hybrid fuzzy neural network to predict the one-day ahead direction of index. We implement different models depending on whether all the indicators and oscillators are used as inputs, or if a linear combination of them obtained through a factor analysis is used instead. In order to guarantee for the robustness of the results, we train and apply the HyFIS models on randomly selected subsamples 10,000 times. The results show that the reduction of the dimension through the factorial analysis generates more profitable and less risky strategies.

Download Full-text

Study on Intelligent Analysis of the Causes of SMT Solder Joint Defects Based on Fuzzy Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.189-193.3257 ◽

2011 ◽

Vol 189-193 ◽

pp. 3257-3261

Author(s):

Chun Yue Huang ◽

He Geng Wei ◽

Tian Ming Li ◽

De Jin Yan

Keyword(s):

Neural Network ◽

Solder Joint ◽

Membership Function ◽

Fuzzy Neural Network ◽

Evaluation Model ◽

Training Samples ◽

Intelligent Analysis ◽

Fuzzy Neural ◽

Input Variables ◽

Set Up

By determining membership function of the input parameters and selecting defuzzification method, the evaluation model which can be used to intelligent analyzing the causes of SMT solder joint defects was set up. The fuzzy neural network was trained by using the output variables of the training samples from intelligent discrimination as the input variables of training samples of fuzzy neural network. The fuzzy neural network was tested by using the output variables of the testing samples from intelligent discrimination as the input variables of testing samples of fuzzy neural network. The results show that by using the evaluation model the cause of SMT solder joint defects can be analyzed intelligently and the results of intelligently analysis are reasonable, the evaluation model can be used practically.

Download Full-text

Fuzzy neural network with support vector-based learning for classification and regression

Soft Computing ◽

10.1007/s00500-019-04116-x ◽

2019 ◽

Vol 23 (23) ◽

pp. 12153-12168 ◽

Cited By ~ 1

Author(s):

Ghazaleh Khodabandelou ◽

Mohammad Mehdi Ebadzadeh

Keyword(s):

Neural Network ◽

Fuzzy Neural Network ◽

Support Vector ◽

Fuzzy Neural ◽

Classification And Regression

Download Full-text

Klasifikasi Motif Batik Berbasis Kemiripan Ciri dengan Wavelet Transform dan Fuzzy Neural Network

ComTech Computer Mathematics and Engineering Applications ◽

10.21512/comtech.v5i1.2630 ◽

2014 ◽

Vol 5 (1) ◽

pp. 361

Author(s):

A Haris Rangkuti

Keyword(s):

Neural Network ◽

Wavelet Transform ◽

Fuzzy Neural Network ◽

Classification Methods ◽

Neural Network Classification ◽

Fuzzy Input ◽

Fuzzy Neural ◽

Input Variables

This paper introduces a classification of the image of the batik process, which is based on the similarity of the characteristics, by combining the method of wavelet transform Daubechies type 2 level 2, to process the characteristic texture consisting of standard deviation, mean and energy as input variables, using the method of Fuzzy Neural Network (FNN). Fuzzyfikasi process will be carried out all input values with five categories: Very Low (VL), Low (L), Medium (M), High (H) and Very High (VH). The result will be a fuzzy input in the process of neural network classification methods. The result will be a fuzzy input in the process of neural network classification methods. For the image to be processed seven types of batik motif is ceplok, kawung, lereng, parang, megamendung, tambal and nitik. The results of the classification process with FNN is rule generation, so for the new image of batik can be immediately known motif types after treatment with FNN classification. For the degree of precision of this method is 86-92%.

Download Full-text

KOMPARASI MODEL SUPPORT VECTOR MACHINES (SVM) DAN NEURAL NETWORK UNTUK MENGETAHUI TINGKAT AKURASI PREDIKSI TERTINGGI HARGA SAHAM

Jurnal Informatika Upgris ◽

10.26877/jiu.v3i1.1536 ◽

2017 ◽

Vol 3 (1) ◽

Author(s):

R. Hadapiningradja Kusumodestoni ◽

Sarwido Sarwido

Keyword(s):

Neural Network ◽

Support Vector Machine ◽

Stock Prices ◽

Prediction Accuracy ◽

High Accuracy ◽

Stock Index ◽

Support Vector ◽

Accuracy Rate ◽

Trend Prediction ◽

Data Set

There are many types of investments to make money, one of which is in the form of shares. Shares is a trading company dealing with securities in the global capital markets. Stock Exchange or also called stock market is actually the activities of private companies in the form of buying and selling investments. To avoid losses in investing, we need a model of predictive analysis with high accuracy and supported by data - lots of data and accurately. The correct techniques in the analysis will be able to reduce the risk for investors in investing. There are many models used in the analysis of stock price movement prediction, in this study the researchers used models of neural networks (NN) and a model of support vector machine (SVM). Based on the background of the problems that have been mentioned in the previous description it can be formulated the problem as follows: need an algorithm that can predict stock prices, and need a high accuracy rate by adding a data set on the prediction, two algorithms will be investigated expected results last researchers can deduce where the algorithm accuracy rate predictions are the highest or accurate, then the purpose of this study was to mengkomparasi or compare between the two algorithms are algorithms Neural Network algorithm and Support Vector Machine which later on the end result has an accuracy rate forecast stock prices highest to see the error value RMSEnya. After doing research using the model of neural network and model of support vector machine (SVM) to predict the stock using the data value of the shares on the stock index hongkong dated July 20, 2016 at 16:26 pm until the date of 15 September 2016 at 17:40 pm as many as 729 data sets within an interval of 5 minute through a process of training, learning, and then continue the process of testing so the result is that by using a neural network model of the prediction accuracy of 0.503 +/- 0.009 (micro 503) while using the model of support vector machine (SVM) accuracy of the predictions for 0477 + / - 0.008 (micro: 0477) so that after a comparison can be concluded that the neural network models have trend prediction accuracy higher than the model of support vector machine (SVM).

Download Full-text

Block Hierarchical Fuzzy-Neural Networks and Their Application to a Mobile Robot Control

Journal of Robotics and Mechatronics ◽

10.20965/jrm.1995.p0012 ◽

1995 ◽

Vol 7 (1) ◽

pp. 12-20

Author(s):

Jun Tang ◽

◽

Keigo Watanabe ◽

Masatoshi Nakamura ◽

◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Mobile Robot ◽

Fuzzy Neural Network ◽

Fuzzy Neural Networks ◽

Small Scale ◽

Circular Path ◽

Single Output ◽

Fuzzy Neural ◽

Input Variables

If some fuzzy sets in a fuzzy-neural network are assigned to each scalar input data, then the number of intermediate unit functions grows exponentially as the number of input variables to the fuzzy reasoning increases. Therefore, it is very important for multi-input/multi-out-put systems to effectively construct a small-scale fuzzy neural network. In this paper, four types of block hierarchical fuzzy-gaussian neural networks (FGNNs) are proposed for a control system of a mobile robot with two independent driving wheels by applying two inputs and single-output FGNN block, or single-input and singleoutput FGNN block. Such a block hierarchical FGNN consists of three layers. In other words, the first input layer consists of two FGNN blocks that independently generate torques for controlling the velocity and azimuth of the mobile robot. The second hidden layer determines their distributions to the final layer by using fixed connection weights. The final output layer also consists of two FGNN bl ks that automatically determine the out put scalers for the actual left- and right-wheel driving torques. The effectiveness of the proposed method is illustrated through some simulations of a circular path tracking control.

Download Full-text