Climate Risk: Explore climate scenario data using
climateScenario object
Create a climateScenario object to perform data exploration, compute new
values, and visualize climate scenario data. You can use the following functions in
your climate scenario workflows:
Market Risk: Use empirical distributions in expected shortfall backtests
Perform expected shortfall (ES) backtests on empirical distributions, such as
historical value-at-risk (VaR) or Monte-Carlo VaR models, by using the
InputData name-value argument when using esbacktestbysim.
Market Risk: Calculate value at risk and expected shortfall
Compute VaR and ES values by using the valueAtRisk and expectedShortfall functions.
Consumer Credit Risk: Validate credit risk models with discrimination and calibration metrics
Validate your credit risk models by using a set of discrimination and calibration metrics. For more information, see:
Binning Explorer app: Generate function to create table of binned data
Binning Explorer now allows you to generate a function that returns a table of binned data. For details, see Binning Explorer.
Binning Explorer app: Export table of binned data
Binning Explorer now allows you to export binned data as a table to your workspace. For details, see Binning Explorer.
Example: Apply granularity adjustment to credit portfolios
This example shows how to apply a granularity adjustment when estimating capital requirements in both small homogeneous credit portfolios and larger nonhomogeneous portfolios. For more information, see Apply Granularity Adjustment to Credit Portfolios.
Modelscape: Use new workflow examples for Modelscape Governance
Use new workflow examples for Modelscape™ Governance™ to manage your financial model inventory and lifecycles with customizable dashboards and workflows. For details, see Modelscape Governance.
Modelscape: Use new examples for Modelscape Validate
Use new features and examples for Modelscape Validate™ to:
Evaluate Modelscape model deployments in the Review Editor. Find different deployments of your models, and call models with a simple evaluation function. For details, see Evaluate Modelscape Deployments with Review Editor.
Use customized signoff forms with large language models (LLMs) to assess model validation findings in Review Editor. For details, see Assess Model Validation Findings with Large Language Models.
Open Python as well as MATLAB® (.m) files in the Review Editor to
inspect model classes and functions written in these languages, with
syntax highlighting. For details, see Analyze Model Version.