Probit Prediction of Going and Non‐going Concerns

Abstract

This article discusses the limitations of existing going‐concern discriminant models and explores the use of weighted probit analysis to construct a classification model for the auditor to use in making going/non‐going concern decisions. The model is constructed using probit analysis with the weighted exogenous sample maximum likelihood (WESML) procedure on a matched sample of 80 going and non‐going concerns. Using the Lachenbruch′s U method of computing holdout accuracy rates, the probit model classifies going/non‐going concerns with an accurate rate of 82.50 per cent for non‐going concerns, 100.00 per cent for going concerns, and 91.50 per cent overall. These accuracy rates are higher than those of the sample auditors, which are 40.00, 97.50, and 68.75 per cent, respectively. The model is expected to be useful as a going‐concern prediction model, a persuasive analytical tool, and a defensive device.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 18 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.