Applied Economics and Finance Vol. 6, No. 1; January 2019 ISSN 2332-7294 E-ISSN 2332-7308 Published by Redfame Publishing URL: http://aef.redfame.com 64 Distinctiveness of Highly Risky Italian Firms That are Saved-A Logistic Approach Marco Muscettola 1 1 Credit Risk Manager-Independent researcher, Adelfia (BA), Italy Correspondence: Marco Muscettola, Credit Risk Manager-Independent researcher, Adelfia (BA), Italy. E-mail: marcomuscettola@hotmail.com Received: November 16, 2018 Accepted: December 3, 2018 Available online: December 7, 2018 doi:10.11114/aef.v6i1.3843 URL: https://doi.org/10.11114/aef.v6i1.3843 Abstract In our paper, we use a default mode approach in order to accurately classify a sample of 3,835 Italian manufacturing companies, and to gauge their health status on the basis of variables taken from the financial statement. The present study is oriented to test the potentiality of salvation for firms included within the worst classes of rating. The research aims to support the resolution of an elaborate theme: the identification of both highly risky companies designed to survive despite their own class of statistical rating, and firms that will move closer to a default status. In this way, the consequences of our examination could help to recognize, among firms considered "highly risky", the latent durability on the time. Keywords: highly risky firms, credit rating, speculative ratings, risk alteration 1. Introduction The present study is oriented to test the potentiality of salvation for firms included within the worst classes of rating. The research aims to support the resolution of an elaborate theme: the identification of both highly risky companies designed to survive despite their own class of statistical rating, and firms that will move closer to a default status. In this way, the consequences of our examination could help to recognize, among firms considered "highly risky", the latent durability on the time. Highly risky companies are for banks the cluster of firms more unmanageable and expensive than other creditworthy companies. That is a serious problem for banks point of view. On the other side, this type of firms is more handy, available and reachable. The study adopts as its starting point a ranking classification to forecast the probability of default built on a three-year period. This time-frame, longer than usual, is innovating compared to the prevalent literature on this matter. Unlike the more common statistical and regulated models to estimate the crisis, in fact, the present revision is intended to be much more forward looking. The essential intention for this paper is to try to perceive latent factors of durability, among a group of firms included in highly risky classes of rating. This characteristic could give banks more information to read between the lines. This study is split in three parts. The first part shows the construction of the rating model to forecast the probability of default and the resulting rating scale. The second part describes the designation of the sub-sample using the rating transition matrix over a period of three years after the first analysis. The sub-sample is composed by the firms with a high risk of default in 2007. The last part, finally, is the discernments of the analysis between the two groups of firms, highly risky in 2007, that have different prospects in 2010, after three years. The subdivision into two sub-samples of analysis also stems from the aspiration to analyse the effects as errors of assessment. In order to construct a prediction model of defaults, using the statistical technique of logistic regression, we test the accuracy of the prevision and we list companies within ten classes of risk. In this way the article focuses on the examination of the eventual distinctions, within the same risk category, among those firms that after three years are still healthy, and those firms that, vice versa, will have become insolvent. In other words, the present dissertation aims to