Statistical associated expected loss, i.e. value of the risk

Statisticalmodels Opposed to approach of Judgmental models is the one of statisticalmodeling, which advocates superiority of quantitative data in establishingunderlying causal relationship between the probability of default and causefactors. Statistical models are governed by statistical methods, consideringmany factors simultaneously, thus calculating and analyzing multivariatecorrelation in order to identify most powerful factors and producestatistically derived weights to be used in consequent scoring model. Useof statistical models in collection process advocates the emphasis placed oninherent risk characteristic of the borrower as opposed to aging items. Why isstatistical model preferential to judgmental in such a case? Reason is that thelatter informs on quality of the risk separating lowest risk accounts fromhighest ones, while statistical model quantifies the risk by informing you onthe probability of default and, therefore, associated expected loss, i.e. valueof the risk (Driving Internal Collection Results With Statistical-based CreditScoring, 2010).  Oneof the most important and prominent contributions made to the field was theobservation by Beaver (1967) that there are several financial ratios, whichdiffer significantly between failed and nonfailed firms, in particular cashflow/net worth and debt/net worth (Falkenstein, Boral, & Carty,RiskCalc(TM) For Private Companies: Moody’s Default Model, 2000). In short,differences in such ratios for viable and bankrupt companies increase as timeto default shortens – as failure neared, firms became more dissimilar.