Model Risk Management with Modelscape

Manage financial models throughout a lifecycle across multiple domains and programming languages

Modelscape™ provides workflow tools for model lifecycle support including governance, automation, documentation, and operation in a unified, customizable system. Use Modelscape if you have a role related to model development and deployment in financial services.

Modelscape provides workflow tools for creating, documenting, validating, deploying, and managing models and associated data. Modelscape supports models you build in MATLAB^®, Python^®, R, SAS^®, and other programming languages.

To host Modelscape for your organization, contact MathWorks Consulting Services. After you have set up Modelscape, to use it in MATLAB R2023a, download the Support Software Downloader for your operating system, run it, and install Modelscape for MATLAB.

Topics

Get Started with Modelscape
Learn the basics of Modelscape.
Modelscape Governance
Manage model inventory, risk scoring, workflows, and reporting.
Modelscape Develop
Develop models with automated documentation and reproducible processes.
Modelscape Validate
Validate models in an interactive environment and perform scenario analysis, issue tracking, and approval tracking.
Modelscape Test
Perform automated preproduction build and test with traceability.
Modelscape Deploy
Host and scale production models in a secure, controlled environment.
Extensibility
Customize Modelscape environment.
Using the Modelscape API
Work programmatically with Modelscape resources such as lifecycles, model versions, and reviews.

Functions

`checkModel`	Check Modelscape model validity (Since R2023a)
`packageModel`	Create Docker image of Modelscape model (Since R2023a)

Apps

Inventory Browser	Create Modelscape models and add dependencies (Since R2023a)
Lifecycle Designer	Create and edit lifecycles in Modelscape (Since R2023a)
Review Editor	Validate Modelscape models and submit reviews (Since R2023a)

Live Editor Tasks

Remove Risk Factors	Remove or include data and record reasons in Modelscape (Since R2021b)
Screen Risk Factors	Remove risk factors from data in Modelscape (Since R2021b)

Featured Examples

Model Development and Experiment Manager

Use MATLAB® Experiment Manager with Modelscape™ at various stages of model development.

Open Live Script

Remove Risk Factors

Remove or include variables from a table and record the corresponding reasons using the Modelscape™ Remove Risk Factors task.

Open Live Script

Fairness Metrics in Modelscape

Detect bias in a credit card data set using a suite of metrics in Modelscape™ software. The metrics are built on the fairnessMetrics object.

Open Live Script

Screen Risk Factors by Custom Criteria

Use the Screen Risk Factors task to automatically exclude risk factors from a table based on their predictive power.

Open Live Script

Model Documentation in Modelscape

Add content to a Microsoft® Word document from MATLAB®.

Open Live Script

Metrics Handlers

Manage Modelscape™ test metrics and their associated threshold objects using MetricsHandler objects.

Open Live Script

Credit Scorecard Validation Metrics

Implement a probability of default (PD) model validation suite covering the techniques laid out in [1]. The techniques in [1] apply to any PD model. In this example, you load the test data, set and compute some key quantities, and calculate some model validation metrics. The results are stored in a collection object. You use the PDModel data set, which consists of a pre-fitted MATLAB® credit scorecard object. You then compute various metrics and summarize them.

Open Live Script