Results (
Thai) 1:
[Copy]Copied!
It is obvious that the accuracy of the model falls off consistently with theone exception of the fourth and fifth years, when the results are reversed fromwhat would be expected. The most logical reason for this occurrence is thatafter the second year, the discriminant model becomes unreliable in its predictiveability, and, also, that the change from year to year has little or nomeaning.Implications. Based on the above results it is suggested that the bankruptcyprediction model is an accurate forecaster of failure up to two years prior tobankruptcy and that the accuracy diminishes substantially as the lead timeincreases. In order to investigate the possible reasons underlying these findingsthe trend in the five predictive variables is traced on a univariate basis for fiveyears preceding bankruptcy. The ratios of four other important but less significantratios are also listed in Table 5.The two most important conclusions of this trend analysis are (1) that allof the observed ratios show a deteriorating trend as bankruptcy approached.
Being translated, please wait..