Credit ratings are an important metric for business managers and a contributor to economic growth. Forecasting such ratings might be a suitable application of big data analytics. As machine learning is one of the foundations of intelligent big data analytics, this chapter presents a comparative analysis of traditional statistical models and popular machine learning models for the prediction of Moody's long-term corporate debt ratings. Machine learning techniques such as artificial neural networks, support vector machines, and random forests generally outperformed their traditional counterparts in terms of both overall accuracy and the Kappa statistic. The parametric models may be hindered by missing variables and restrictive assumptions about the underlying distributions in the data. This chapter reveals the relative effectiveness of non-parametric big data analytics to model a complex process that frequently arises in business, specifically determining credit ratings.
|Title of host publication||Managerial Perspectives on Intelligent Big Data Analytics|
|Number of pages||19|
|ISBN (Print)||9781522572770, 1522572775|
|Publication status||Published - Feb 2019|