SFFS-PC-NN optimized by genetic algorithm for dynamic prediction of financial distress with longitudinal data streams

Application of Meta-Heuristic Algorithms for Training Neural Networks and Deep Learning Architectures: A Comprehensive Review

The learning process and hyper-parameter optimization of artificial neural networks (ANNs) and deep learning (DL) architectures is considered one of the most challenging machine learning problems. Several past studies have used gradient-based back propagation methods to train DL architectures. However, gradient-based methods have major drawbacks such as stucking at local minimums in multi-objective cost functions, expensive execution time due to calculating gradient information with thousands of iterations and needing the cost functions to be continuous. Since training the ANNs and DLs is an NP-hard optimization problem, their structure and parameters optimization using the meta-heuristic (MH) algorithms has been considerably raised. MH algorithms can accurately formulate the optimal estimation of DL components (such as hyper-parameter, weights, number of layers, number of neurons, learning rate, etc.). This paper provides a comprehensive review of the optimization of ANNs and DLs using MH algorithms. In this paper, we have reviewed the latest developments in the use of MH algorithms in the DL and ANN methods, presented their disadvantages and advantages, and pointed out some research directions to fill the gaps between MHs and DL methods. Moreover, it has been explained that the evolutionary hybrid architecture still has limited applicability in the literature. Also, this paper classifies the latest MH algorithms in the literature to demonstrate their effectiveness in DL and ANN training for various applications. Most researchers tend to extend novel hybrid algorithms by combining MHs to optimize the hyper-parameters of DLs and ANNs. The development of hybrid MHs helps improving algorithms performance and capable of solving complex optimization problems. In general, the optimal performance of the MHs should be able to achieve a suitable trade-off between exploration and exploitation features. Hence, this paper tries to summarize various MH algorithms in terms of the convergence trend, exploration, exploitation, and the ability to avoid local minima. The integration of MH with DLs is expected to accelerate the training process in the coming few years. However, relevant publications in this way are still rare.

Dynamization of bankruptcy models via indicator variables

Purpose

The purpose of this paper is to enhance the predictive power of bankruptcy prediction models by taking the past values of firms’ financial ratios as benchmark. For this purpose, the paper proposes an indicator variable expressing the time trends of financial ratios.

Design/methodology/approach

The proposed measure uses the minimum and the maximum of financial ratios from the previous period as benchmarks in order to give a more complete picture about the present financial performance of firms. The most popular classification methods of bankruptcy prediction were employed: discriminant analysis, logistic regression, decision trees. Sample specific results and conclusions were avoided by applying tenfold stratified cross-validation.

Findings

The empirical results suggest that the proposed measure can increase the predictive performance of bankruptcy prediction models compared to models based solely on static financial ratios. The results gave evidence for the fact that the firms’ past financial performance is a useful benchmark for evaluating the risk of future insolvency.

Originality/value

The proposed concept is completely new to the literature and practice of bankruptcy prediction. Similar concept has not been published to date. The suggested dynamization approach has three important advantages. It is easy to compute from time series of financial ratios. It is applicable within any classifier irrespective of its mathematical background. The performance of models can be enhanced without the necessity of giving up the interpretability of bankruptcy models, so the proposed measure may play very important role in the practice of credit scoring modeling as well.

Improving the prediction of going concern of Taiwanese listed companies using a hybrid of LASSO with data mining techniques

The purpose of this study is to establish rigorous and reliable going concern doubt (GCD) prediction models. This study first uses the least absolute shrinkage and selection operator (LASSO) to select variables and then applies data mining techniques to establish prediction models, such as neural network (NN), classification and regression tree (CART), and support vector machine (SVM). The samples of this study include 48 GCD listed companies and 124 NGCD (non-GCD) listed companies from 2002 to 2013 in the TEJ database. We conduct fivefold cross validation in order to identify the prediction accuracy. According to the empirical results, the prediction accuracy of the LASSO–NN model is 88.96 % (Type I error rate is 12.22 %; Type II error rate is 7.50 %), the prediction accuracy of the LASSO–CART model is 88.75 % (Type I error rate is 13.61 %; Type II error rate is 14.17 %), and the prediction accuracy of the LASSO–SVM model is 89.79 % (Type I error rate is 10.00 %; Type II error rate is 15.83 %).