Adversarial Semi-supervised Learning for Corporate Credit Ratings

Corporate credit rating is an analysis of credit risks withina corporation, which plays a vital role during the management of financial risk. Traditionally, the rating assessment process based on the historical profile of corporation is usually expensive and complicated, which often takes months. Therefore, most of the corporations, duetothelack in money and time, can’t get their own credit level. However, we believe that although these corporations haven’t their credit rating levels (unlabeled data), this big data contains useful knowledgeto improve credit system. In this work, its major challenge lies in how to effectively learn the knowledge from unlabeled data and help improve the performance of the credit rating system. Specifically, we consider the problem of adversarial semi-supervised learning (ASSL) for corporate credit rating which has been rarely researched before. A novel framework adversarial semi-supervised learning for corporate credit rating (ASSL4CCR) which includes two phases is proposed to address these problems. In the first phase, we train a normal rating system via a machine-learning algorithm to give unlabeled data pseudo rating level. Then in the second phase, adversarial semi-supervised learning is applied uniting labeled data and pseudo-labeleddatato build the final model. To demonstrate the effectiveness of the proposed ASSL4CCR, we conduct extensive experiments on the Chinese public-listed corporate rating dataset, which proves that ASSL4CCR outperforms the state-of-the-art methods consistently.

Identification of best discrimination surface by mixed-integer semi-definite programming for support vector machine

This paper proposes two improvements to the support vector machine (SVM): (i) extension to a semi-positive definite quadratic surface, which improves the discrimination accuracy; (ii) addition of a variable selection constraint. However, this model is formulated as a mixed-integer semi-definite programming (MISDP) problem, and it cannot be solved easily. Therefore, we propose a heuristic algorithm for solving the MISDP problem efficiently and show its effectiveness by using corporate credit rating data.

The Regulatory Role of Credit Ratings and Voluntary Disclosure

We find that corporate credit rating changes have an effect on firms' voluntary disclosure behavior that is independent of the information they convey about firm fundamentals. Our analyses exploit two separate quasi-experimental settings that generate either exogenous credit rating downgrades or credit rating upgrades (i.e., credit rating label changes). We find evidence of a negative relation between the direction of the credit rating label change and the provision of voluntary disclosure in both settings-firms respond to exogenous downgrades by increasing voluntary disclosure and to exogenous upgrades by decreasing voluntary disclosure. The effects we document are attributable to the regulatory role rather than the information role of credit ratings. Overall, our analyses indicate that credit rating agencies as gatekeepers influence firms' provision of voluntary disclosure.

A Multi-Country Study of Factors Affecting Credit Ratings Revisions

This paper reveals findings from extending corporate credit rating studies towards (i) new ratings, affirmation, confirmation, watchlists, and withdrawal, which together represent five out of eight rating types yet to be studied rigorously (there are several papers on upgrades and downgrades); and (ii) identifying key firm-specific factors affecting stock prices around the rating revisions in markets not yet studied. The firmspecific factor effects are measured using the Ordered Probit methodology. Results show that investment and speculation grade issues have the most pronounced effects on price changes. Further findings are: interest-coverage, profitability and leverage ratios, all of which stand out as the most relevant firm-specific factors correlated with stock price changes. An interesting new finding is the discovery of corruption perception scores as a new measure is significantly influencing affirmation, confirmation and downgrade ratings. These new findings are likely to be of interest to investors, corporations wanting to know rating change effects and the external regulators concerned with financial weaknesses/strengths of listed firms facing rating changes.

Predicting corporate credit rating based on qualitative information of MD&A transformed using document vectorization techniques

PurposeThe purpose of this study is to investigate the effectiveness of qualitative information extracted from firm’s annual report in predicting corporate credit rating. Qualitative information represented by published reports or management interview has been known as an important source in addition to quantitative information represented by financial values in assigning corporate credit rating in practice. Nevertheless, prior studies have room for further research in that they rarely employed qualitative information in developing prediction model of corporate credit rating.Design/methodology/approachThis study adopted three document vectorization methods, Bag-Of-Words (BOW), Word to Vector (Word2Vec) and Document to Vector (Doc2Vec), to transform an unstructured textual data into a numeric vector, so that Machine Learning (ML) algorithms accept it as an input. For the experiments, we used the corpus of Management’s Discussion and Analysis (MD&A) section in 10-K financial reports as well as financial variables and corporate credit rating data.FindingsExperimental results from a series of multi-class classification experiments show the predictive models trained by both financial variables and vectors extracted from MD&A data outperform the benchmark models trained only by traditional financial variables.Originality/valueThis study proposed a new approach for corporate credit rating prediction by using qualitative information extracted from MD&A documents as an input to ML-based prediction models. Also, this research adopted and compared three textual vectorization methods in the domain of corporate credit rating prediction and showed that BOW mostly outperformed Word2Vec and Doc2Vec.