search query: @indexterm FORECASTING TECHNIQUES / total: 246
reference: 51 / 246
Author: | Koh, H. C. |
Title: | The sensitivity of optimal cutoff points to misclassification costs of Type I and Type II errors in the going-concern prediction context. |
Journal: | Journal of Business Finance and Accounting
1992 : JAN, VOL. 19:2, p. 187-197 |
Index terms: | FORECASTING TECHNIQUES FINANCIAL MODELS LOGIT MODELS |
Language: | eng |
Abstract: | To investigate the sensitivity of optimal cutoff points to misclassification costs of Type I and Type II errors, a going-concern prediction model was first constructed using logit analysis on a matched sample of 165 going concerns and 165 non-going concerns. Next, optimal cutoff points were determined using different relative misclassification costs ranging from 1:1 to 500:1. The results show that optimal cutoff points in going-concern prediction models are insensitive to different relative misclassification costs. Further, going-concern prediction models can achieve high accuracy rates. Thus, they can be used to aid auditors in making going-concern assessments and preliminary reviews. |
SCIMA