A Seasonal Time-Series Model Based on Gene Expression Programming for Predicting Financial Distress

Table 1

Financial crisis related researches for research methods and results.

Author	Research methods	Data period	Sample ratio (crisis : health)	Accuracy
Frydman et al. [17]	Repeated segmentation logic	1971–1981	58 : 142	85.0–94.0
Mossman et al. [18]	LR	1970–1976	Moody’s Industrial Manual 23 : 23	82.6–84.9
Atiya [19]	NN	1993–1996	US 444 : 716	60.0–90.8
Chen [20]	PCA + DT, PCA + LR	2000–2007	Taiwan 50 : 50	85.1–97.0
Li et al. [21]	RSBL, MDA, Logit, Probit	N/A	China 135 : 135,	71.6–88.5
Korol [2]	MDA, DT, NN	1993–1996	Warsaw 50 : 135; Latin America 60 : 30	74.1–96.8
Geng et al. [22]	DT, SVM, NN	Open dataset	China 344 : 344; Taiwan 220 : 220; Australian 307 : 382; German 700 : 300	70.9–92.1
Liang et al. [23]	SVM, RBF SVM, -NN, NB, CART, MLP	Open dataset	China 344 : 344; Taiwan 220 : 220; Australian 307 : 382; German 700 : 300	70.9–92.1

MDA: multivariate discriminant analysis, DT: decision tree, NN: neural network, PCA: principal component analysis, LR: logistic regression, NB: Naive Bayes, MLP: multilayer perceptron neural network, CART: classification and regression tree, SVM: support vector machines, and RSBL: random subspace binary logit.