José Willer do Prado

Multivariate analysis of credit risk and bankruptcy research data: a bibliometric study involving different knowledge fields (1968---2014)

Bibliometrics are important evaluation tools to the scientific production. A bibliometric study was used to evaluate researches about credit risk and bankruptcy. Credit support technique studies on economic and social orders are relevant in this knowledge field. Therefore, the aim of the current study is to identify and describe the application of multivariate data analysis techniques to credit risk and bankruptcy scenarios. The herein presented data were collected in publications indexed to Thomson Reuters’ Web of Science database between 1968 and 2014. The results corroborate information in the literature and in previous bibliometric reviews, as well as highlight other indications regarding the construction and development of research fields. Since the 1990’s, the neural networks became relevant due to their increased use as study object in publications. However, both the discriminant analysis (J Finance 23(4):589–609, 1968. doi:10.2307/2978933) and the logistic regression (J Account Res 18(1):109–131, 1980. doi:10.2307/2490395) are still often used in researches, fact that shows the tendency to find articles using more than one technique or hybrid models, artificial intelligence techniques and complex computer systems. This field appears to be multidisciplinary in journals and Web Science databases involving the business and economy, operational research, management, mathematics, data processing, engineering and statistics fields. Another relevant discovery was the increased number of publications about this subject launched right after the 2008 crisis.

Bibliometric analysis on the association between behavioral finance and decision making with cognitive biases such as overconfidence, anchoring effect and confirmation bias

The aim of the present study is to conduct a bibliometric analysis on the association between the themes ‘behavioral finance’ and ‘financial and managerial decision making’, and the cognitive biases ‘overconfidence’, ‘anchoring effect’ and ‘confirmation bias’. The search for articles was performed at the Web of Science database using EndNote® as reference management software, and CiteSpace (Chen in Proc Natl Acad Sci 101(suppl 1):5303–5310, 2004; J Am Soc Inf Sci Technol 57(3):359–377, 2006) as bibliometric analysis software. The search led to 889 articles published between 1990 and 2016, and the results have shown that the number of researches relating overconfidence, anchoring and confirmation biases to behavioral finances has been growing throughout time, mainly from 2008 on. Besides, the results have confirmed the importance of Amos Tversky and Daniel Kahneman to this research field. The bias presenting the closest proximity to the behavioral finance field in the present study was overconfidence. The confirmation bias was the one presenting the smallest number of publications and the slightest relation to this study field, fact that opens a promising research field.