Credolab is at the forefront of innovative risk management practices that engage with novel credit risk modelling approaches availed by the surge in cell phone use. Core to credolab’s business is its modelling pipeline. Taking the smartphone as input, the data processing pipeline consists of a series of automated steps, rooted in machine learning techniques, that ultimately outputs a predictive model for credit default. To protect the confidentiality and to ensure against bias towards individual loan customers, only non-identifying metadata is used.
This e-book reports the findings of Dr Xiaofei (Susan) Wang, Lecturer and Research Scholar, Yale University from a review she did on credolab’s scoring model. She considered a vast array of alternative approaches for the various different steps of the pipeline and found favourable results, including when applied to real data.
In this e-book, we first explore the data sets that credolab consumes, how it translates it into scores, and the outcome it serves. In the latter part of the paper, we take a look at how credolab's algorithm fared when compared to that of other major players with similar scoring models.