Neural Network design parameters for forecasting financial time series

Lasfer, Assia; El-Baz, Hazim; Zualkernan, Imran A.

doi:10.1109/icmsao.2013.6552553

Cited by 9 publications

(10 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, we have prepared a literature review where genetic algorithms and design of experiments are used as an optimization tool in exchange rate prediction [21]. The most common DOE methods that are used for hyperparameter optimization in time series prediction are Taguchi method [22], response surface methodology [23], full factorial design [24] etc. In addition to the evolutionary and statistical approaches, multi-criteria decision making techniques are also used for hyperparameter optimization in time series prediction.…”

Section: Literature Reviewmentioning

confidence: 99%

Design of Experiments vs. TOPSIS to Select Hyperparameters of Neural Attention Models in Time Series Prediction

Midilli

Parshutin

2020

ITMS

View full text Add to dashboard Cite

Attention models are used in neural machine translation to overcome the challenges of classical encoder-decoder models. In the present research, design of experiments and TOPSIS methods are used to select hyperparameters of a neural attention model for time series prediction. The configurations selected by both methods are compared with out-of-sample data in time interval between January 2020 and April 2020 when global economies were significantly impacted due to Covid-19 pandemic. Results demonstrated that both selection methods outperformed each other in terms of different output features. On the other hand, our results with more than 95 % coefficient of determination and less than 0.23 % MAPE verified that neural attention models had strong capabilities in exchange rate prediction even in extraordinary situations in global economies.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Design of Experiments vs. TOPSIS to Select Hyperparameters of Neural Attention Models in Time Series Prediction

Midilli

Parshutin

2020

ITMS

View full text Add to dashboard Cite

show abstract

“…The performance of DNN is highly influence by the its parameters for a particular domain. Lasfer el at [146] performed an analysis on the influence of parameter (like the number of neurons, learning rate, activation function etc) on stock price forecasting. The authors work showed that a larger NN produces a better result than a smaller NN.…”

Section: B Financial Time Series Forecastingmentioning

confidence: 99%

“…Although CNN is well known for its stripes in image recognition and less application in the Financial markets, CNN have also shown good performance in forecasting the stock market. As indicated by [146], the deeper the network the more NN can generalize to produce good results. However, the more the layers of NN, it is more likely to overfit a given data set.…”

Section: B Financial Time Series Forecastingmentioning

confidence: 99%

A Review of Deep Learning with Special Emphasis on Architectures, Applications and Recent Trends

Sengupta¹,

Basak²,

Saikia³

et al. 2019

Preprint

View full text Add to dashboard Cite

Deep learning has taken over - both in problems beyond the realm of traditional, hand-crafted machine learning paradigms as well as in capturing the imagination of the practitioner sitting on top of petabytes of data. While the public perception about the efficacy of deep neural architectures in complex pattern recognition tasks grows, sequentially up-to-date primers on the current state of affairs must follow. In this review, we seek to present a refresher of the many different stacked, connectionist networks that make up the deep learning architectures followed by automatic architecture optimization protocols using multi-agent approaches. Further, since guaranteeing system uptime is fast becoming an indispensable asset across multiple industrial modalities, we include an investigative section on testing neural networks for fault detection and subsequent mitigation. This is followed by an exploratory survey of several application areas where deep learning has emerged as a game-changing technology - be it anomalous behavior detection in financial applications or financial time-series forecasting, predictive and prescriptive analytics, medical imaging, natural language processing or power systems research. The thrust of this review is on outlining emerging areas of application-oriented research within the deep learning community as well as to provide a handy reference to researchers seeking to embrace deep learning in their work for what it is: statistical pattern recognizers with unparalleled hierarchical structure learning capacity with the ability to scale with information.

show abstract

“…It was concluded that while traditional models had 53-64 % accuracy, the proposed model had 84 % accuracy. The study [15] identified the design parameters that might impact NN built for forecasting future moves of the UAE MSCI Index in the time interval between 2002 and 2012. Two-level full factorial DOE was established for the design parameters: NN type, number of neurons in hidden layers, output layer transfer function and learning rate.…”

Section: Optimisation Of Nn Hyper-parametersmentioning

confidence: 99%

“…There are still concerns and challenges to optimise NNs in the financial data analysis. NN design is done through calibrating numerous parameters; however, finding the correct combination of parameter values is challenging given a specific data set [14], [15]. In general, the trial and error process is what most practitioners use for optimisation [16].…”

Section: Introductionmentioning

confidence: 99%

Review for Optimisation of Neural Networks With Genetic Algorithms and Design of Experiments in Stock Market Prediction

Midilli

Parshutin

2019

ITMS

View full text Add to dashboard Cite

Neural networks are commonly used methods in stock market predictions. From the earlier studies in the literature, the requirement of optimising neural networks has been emphasised to increase the profitability, accuracy and performance of neural networks in exchange rate prediction. The paper proposes a literature review of two techniques to optimise neural networks in stock market predictions: genetic algorithms and design of experiments. These two methods have been discussed in three approaches to optimise the following aspects of neural networks: variables, input layer and hyper-parameters.

show abstract

Neural Network design parameters for forecasting financial time series

Cited by 9 publications

References 7 publications

Design of Experiments vs. TOPSIS to Select Hyperparameters of Neural Attention Models in Time Series Prediction

Design of Experiments vs. TOPSIS to Select Hyperparameters of Neural Attention Models in Time Series Prediction

A Review of Deep Learning with Special Emphasis on Architectures, Applications and Recent Trends

Review for Optimisation of Neural Networks With Genetic Algorithms and Design of Experiments in Stock Market Prediction

Contact Info

Product

Resources

About