Navigation :
test ../ test user-ref.html
User & Reference Guide
test ../ test getting-started.html
Getting started
test ../ test getting-started/using-this-guide.html
- Using this guide
test ../ test getting-started/whats-new.html
- What's New
test ../ test getting-started/frtb-data-model.html
- FRTB Data Model
test ../ test tutorials.html
Tutorials
test ../ test tutorials/data-health-check.html
- Data Health Check
test ../ test tutorials/tips-for-validating-the-calculations.html
- Tips for Validating the Calculations
test ../ test tutorials/viewing-qis-numbers.html
- Viewing QIS Numbers
test ../ test tutorials/workaround-for-deadlock-issue-when-attempting-to-create-partitions-for-reference-stores.html
- Workaround for deadlock issue when attempting to create partitions for reference stores
test ../ test configuration.html
Configuration files
test ../ test sign-off.html
Sign-Off Approvals
test ../ test cube.html
Data Cubes
test ../ test cube/measures.html
-
Measures
test ../ test cube/dimensions.html
-
Dimensions
test ../ test what-if.html
What-If Analysis
test ../ test datastore.html
Data Stores
test ../ test input-files.html
Input Files
test ../ test dev.html
Developer Guide
test ../ test dev/dev-release.html
-
Release and migration notes
test ../ test dev/dev-getting-started.html
-
Getting Started
test ../ test dev/dev-ref-impl.html
-
FRTB Reference Implementation
test ../ test dev/dev-core.html
-
FRTB Core
test ../ test dev/dev-extensions.html
-
Extending the Accelerator
test ../ test dev/dev-tools.html
-
Configuring Accelerator tools and methodologies
test ../ test dev/dev-sign-off.html
-
Sign-Off
test ../ test interpret-impl.html
FRTB Accelerator Interpretation and Implementation of BCBS 457
test ../ test pdf-guides.html
PDF Guides
test ../ test glossary.html
Glossary
DifferentiationStepSize
Description
Size of the step in numerical approximation calculation
Variations
Context value DifferentiationStepSize specifies the size of the step in numerical derivative approximation. This applies to the numerical Euler calculations only. The default value is close to 1E-8 (i.e. 10-8 or 0.00000001).