| This study assessed the performance of predictions model when applied to certain Standard Industrial Classifications (SIC); primarily manufacturing, transportation, and trading, compared with the model's performance when applied without regard to industry (across SIC).; Samples in the studies consisted of 118 bankrupt firms filing bankruptcy during the period of 1975-1983, and 118 nonbankrupt firms existing in 1983. Among 118 firms in each group, 69 were in manufacturing, 17 in transportation, and 32 in trading industries.; The prediction model was derived from financial variables of bankrupt and nonbankrupt firms using discriminant analysis. The discriminant function was constructed first from the sample across SIC, then, from manufacturing, transportation, and trading industries. The prediction accuracy for a five year period prior to bankruptcy and the contribution of variables in the model were determined.; Based on the results of the study, predictions of bankruptcy based on industry differences performed significantly better than predictions without regard to industry for manufacturing and transportation industries. Also, it was found that the set of variables that predicted some SIC groups successfully could not predict other SIC groups with the equivalent successful level. |
