roboR8 is a corporate credit rating system based upon a new approach in Machine Learning (ML) which combines deep learning with optimal classification trees.

The roboR8 methodology takes a set of financial ratios, together with macro-economic data, to create the system inputs. From these inputs, the model optimally chooses the subset which is most relevant and the chosen subset is then used as input into a machine-learning classifier which creates a non-linear mapping from the input to the output (the rating). The inputs have no reliance upon traded instruments and, therefore, the same methodology can be applied to public, private, large or small companies.

Our Machine Intelligence via Graph Optimization (MIGO) classifier that performs the above mapping is proprietary and is based on an optimal** **algorithm developed by our research team. The superiority of MIGO over other existing ML algorithms has been demonstrated both theoretically and in practice on this and several other applications.

The roboR8 system using the MIGO rating generator produces a **Rating Distribution **rather than a simple rating. The full rating probability distribution of possible ratings can be used to:

- Assess the confidence of the assigned rating
- Present more detailed risk insight by giving a probability of a lower/higher rating along with rating transition probabilities
- Compare same-rated corporates: companies with the same single-label rating do not possess the same risk and the distributional rating output can be used to make an objective credit comparison between companies

No dependence upon analysts and therefore an objective rating without human bias or subjectivity

The individual classification trees produced are theoretically proven to be optimal, thus outperforming any classification tree derived from the same factor-set

Due to the algorithm being optimal, rather than a heuristic, MIGO requires smaller amounts of data than competing systems to produce accurate results

A single forest of models is constructed for each industry and region. Model consistency is therefore assured across region, industry, size, public and private sectors

MIGO deploys a computationally efficient algorithm and places no restriction upon the volume of issuers or the frequency of rating updates

Unlike some other machine learning algorithms, (which may use an opaque “black-box” approach to forecasting/classification) MIGO allows the user to gain insight into the dependence of the rating on the components of the financial statement and macro-economic factors, therefore, naturally suggesting scenario analyses

The model makes no assumption about when the target-labels are observed and hence, can be used to forecast a rating an arbitrary horizon window

The roboR8 rating is congruent to existing rating agencies and can be translated to their scale, thus allowing a direct comparison

MIGO is a semi-supervised system and if target data is sparse, cannot be used, or is non-existent, then MIGO uses similarity algorithms and knowledge transference to get rating results

To view the roboR8 system, please click below. For user registration, please send us an email

Copyright © 2020 Rekiki PTE Ltd, 30 Cecil Street, #19-08 Prudential Tower, Singapore 049712. UEN 201013123D