Comparison of new activation functions in neural network for forecasting financial time series

Authors:
Gecynalda S. da S. Gomes;Teresa B. Ludermir;Leyla M. M. R. Lima
Affiliations:
Centre of Informatics, Federal University of Pernambuco, Av. Prof. Luis Freire, s/n, Cidade Universitria Recife/PE, 50732-970, Recife, Brazil;Centre of Informatics, Federal University of Pernambuco, Av. Prof. Luis Freire, s/n, Cidade Universitria Recife/PE, 50732-970, Recife, Brazil;Centre of Informatics, Federal University of Pernambuco, Av. Prof. Luis Freire, s/n, Cidade Universitria Recife/PE, 50732-970, Recife, Brazil
Venue:
Neural Computing and Applications
Year:
2011

Citing 0
Cited 4

Modified wavelet neural network in function approximation and its application in prediction of time-series pollution data

Applied Soft Computing
Objective investigation of vision impairments using single trial pattern reversal visually evoked potentials

Computers and Electrical Engineering
Optimization of the weights and asymmetric activation function family of neural network for time series forecasting

Expert Systems with Applications: An International Journal
A comparative investigation of non-linear activation functions in neural controllers for search-based game AI engineering

Artificial Intelligence Review

Quantified Score

Hi-index	0.00

Visualization

Abstract

In artificial neural networks (ANNs), the activation function most used in practice are the logistic sigmoid function and the hyperbolic tangent function. The activation functions used in ANNs have been said to play an important role in the convergence of the learning algorithms. In this paper, we evaluate the use of different activation functions and suggest the use of three new simple functions, complementary log-log, probit and log-log, as activation functions in order to improve the performance of neural networks. Financial time series were used to evaluate the performance of ANNs models using these new activation functions and to compare their performance with some activation functions existing in the literature. This evaluation is performed through two learning algorithms: conjugate gradient backpropagation with Fletcher–Reeves updates and Levenberg–Marquardt.