Table 4.

The similar experiments were carried out with FNN ANFIS.

Total amount of errors 5 % of errors 10 First type of errors 0 Second type of errors 5

Accounting and Finance - New Perspectives on Banking, Financial Statements and Reporting

The results of application of FNN TSK are presented in Table 2.

Total amount of errors 6 % of errors 12 First type of errors 1 Second type of errors 5

The similar experiments were carried out with FNN ANFIS.

Total number of errors

Comparative analysis of FNN ANFIS and TSK in dependence on rules number.

The comparative analysis of forecasting results versus the number of rules is

Number of first type errors

Number of second type errors

% of errors

ANFIS 5 6 12 0 6 ANFIS 10 7 14 1 6 TSK 5 5 10 0 5 TSK 10 6 12 1 5

Comparing the results in Table 3, one may conclude FNN TSK has better

The goal of the next experiments was to explore the influence of training and

Training sample—120 Ukrainian banks, test sample—50 banks, and number of

The goal of the next experiment was to find out the dependence of rule number on predicting accuracy. Input data—the same financial indices as in experiment 1.

Experiment No. 2:

Results

Table 2.

Table 3.

Results of FNN TSK forecasting.

Results

Table 1.

Experiment No. 3:

presented in Table 3 [4].

Network/number of rules

Results of FNN TSK forecasting.

accuracy than FNN ANFIS.

Experiment No. 4:

rules = 10.

88

test samples size on accuracy of forecasting.

Results of FNN TSK forecasting.

The results for FNN TSK are presented in Table 4. The similar experiments were carried out with FNN ANFIS. After analysis of the experimental results the following conclusions were made:

FNN TSK ensures the higher accuracy of risk forecasting than FNN ANFIS;

the variation of the number of rules in the training and test samples makes slight influence on the accuracy of forecasting; and

the goal of the next series of experiments was to determine the optimal input data (financial indicators) for bankruptcy risk forecasting. The period of input data was January 2008.
