EN
keyboard_arrow_down
Choose a language:

Model Validation Engine

An automated platform for validation and monitoring of risk models

How does it work

Model Validation Engine is an innovative application that automates and facilitates the process of model validation. Navigation in a user-friendly interface allows for accurate model quality analysis, real-time verification of calculation results and report generation in the form of a structured and editable document.

What challenges does the Model Validation Engine address?

  • Model risk management is one of the major challenges any financial institution is facing. In the face of increasing requirements of regulators emphasizing the need for continuous model quality verification and decreasing availability of highly qualified specialists in the market, it is necessary to introduce a significant degree of automation of this process.
  • The Validation Engine model has been designed to automate validation and monitoring tasks that are repetitive and very time consuming. Our application ensures that model risk management teams can focus not on manually calculating the necessary statistics, but on analyzing the results of calculations and drawing conclusions for the future.

 

4 easy steps to perform validation with Model Validation Engine

01

Upload the data

Model Validation Engine accepts input data for validation in a form of either a flat file, Excel or SQL database. The Validation Engine model does not impose any data structure. The user uploads the input file and then defines the content of the dataset.

02

Set acceptance criteria

The user can define their own traffic-light acceptance criteria for all validation tests which can be saved and used in the subsequent validation exercises.

03

Execute the tests

Perform selected tests and analyze the results in real time through the application’s interface. In addition to exact numerical results, the outcome is visualized on insightful charts.

04

Generate the report

Once the model validation process is completed, the results can be exported to a well-structured editable document (Word, Excel etc.) which can be directly used as a part of model validation or monitoring documentation.

Benefits

autorenewAutomation

Model Validation Engine automates repetitive monitoring and validation activities allowing the team to dedicate time to exploring new areas of risk modelling.

mouseOne-click reporting

Automated report generation makes the final results available in less than a minute. It allows the user to recreate the documentation multiple times, due to changing data, parameterization or segmentation, without any additional time or effort.

assignmentMethodology standardization

Model Validation Engine is equipped with validation tests compliant with the best market practice and external (supervisory) requirements, in order to provide a unified framework of model testing.

searchProcess auditability

Model Validation Engine structures the entire validation/monitoring process by assigning users appropriate roles and access rights. It also provides full auditability of the results and all validation/monitoring performed (version control, business and technical logs), which significantly reduces operational risk.

Mariusz Śpiewak

Both supervisors and the market in general expect significantly higher standards regarding data quality and mathematical correctness of risk models as well as their documentation. The requirements regarding frequency of model testing are also increasing. This creates space where modern technology can make a difference and Model Validation Engine is our answer to these challenges.

Mariusz Śpiewak, Partner at PwC Polska and Consulting Leader

Pricing

On-premise implementation
The implementation price depends on the scope of the business configuration

Components

  • One-time implementation fee
    1. Model Validation Engine implemented on the customer's infrastructure (authentication, user authentication and the ability to connect to internal data sources)
    2. Configuration of the Validation Engine Model to the client's needs in terms of the validation tests used
    3. Training
  • Annual fee
    1. Maintenance (10% of the implementation fee) - handling of patches, required changes to the code, diagnostics and technical support, software versioning and updating, etc.
    2. License (10% of the implementation fee)

Contact us

Do you want to talk about how our tool will help you automate and simplify the model validation process?

Leave your details below to contact our experts.

Fields marked with * are mandatory.
By submitting your email address, you confirm that you have read the Privacy Policy and consent to the processing of data in accordance with the Privacy Policy (including international transfers). If you change your mind about receiving information from us, you can send us an e-mail


You may also like

.