An assessment of pension liabilities according to IAS 19 «Employee Benefits» and national standards tab_img
Recultivation reserves: assessment of provisions according to IAS 37 (ARO, ERL etc.) tab_img
Machine learning and predictive modelling tab_img
Non-life insurance tab_img
Life insurance tab_img

Machine learning and predictive modelling

An increasingly large number of industrial and financial companies start using methods of machine learning and data science to reveal hidden characteristics and forecast client’s behaviour.

Our developments have been successfully implemented in insurance organisations, credit alliances and other financial institutions. Our solutions in this area allow our clients to utilize a potential of accumulated data to the full extent.


The up-to-date methods of the insurers’ data analysis make it possible to:

  • Detect factors affecting the frequency and amount of insurance losses and their interrelation.
  • Select insurance objects more precisely, determining their value and manage the insurance portfolio.
  • Decrease the loss ratio and improve operational results.
  • Improve client’s profitability for business purposes relying on a long-term forecast of client’s behaviour, etc.

The range of services may change depending on the client’s needs.

Based on the data provided by banks and credit institutions, we develop models that enable us to:

  • Assess and forecast a default risk (scoring cards).
  • Detect and prevent fraud.
  • Build a good strategy of recovering overdue indebtedness.
  • Calculate the credit value subject to all costs and expenses, and the future events forecast.

The range of services may change depending on the client’s needs.

Implementing up-to-date methods of data analysis allows our clients to improve key performance indicators and become more competitive.