CiteWeb id: 20160000011

CiteWeb score: 1641

Researchers typically estimate financial distress prediction models on nonrandom samples. Estimating models on such samples can result in biased parameter and probability estimates if appropriate estimation techniques are not used. This paper examines conceptually and empirically two estimation biases which can result when financial distress models are estimated on nonrandom samples. The first bias results from "oversampling" distressed firms and falls within the topic of choice-based sample biases. The second results from using a "complete data" sample selection criterion and falls within the topic of sample selection biases. The two issues examined in this paper arise because of sample selection/data collection constraints typically faced by financial distress researchers. The first constraint is the extremely low frequency rate of firms exhibiting financial distress characteristics (e.g., petitioning for