Financial risk analytics and modeling

Services and products for every area of risk analytics, modeling and model risk management

Overview

Use of models and broader analytics in the financial services industry continues to expand at a rapid pace and penetrate ever broader set of business uses and processes. We are also seeing greater reliance on more complex machine learning techniques and availability of industry tools employing them.

PwC’s risk analytics professionals are dedicated to bringing our clients a combination of qualitative, quantitative, and technology skills, along with over 20 years of deep knowledge of regulatory expectations and leading industry practices in modeling and analytics.

The financial risk analytics and modeling lifecycle

Our analytical products and services cover the full model lifecycle and the entire spectrum of business and functional areas.

Model governance

Since the inception of the model risk management (MRM) discipline in 2000 with the introduction of the OCC 2000-16 guidance, the MRM practices have been evolving to adapt to ever greater reliance on models by institutions. PwC’s MRM professionals have been on the forefront of these developments, helping drive thought leadership and practices for effective and efficient model risk management.

How can we help?

Build an effective model risk management framework

PwC’s model risk management specialists can help your institution build a comprehensive model risk management framework or enhance elements of your existing framework. Our library of MRM policies, procedures, standards, templates and tools developed for institutions of different size and complexity allows us to tailor a framework that works well for your organization.
We can help enhance your MRM framework to help better manage the risks associated with the use of artificial intelligence / machine learning models.

Provide model risk management technology platform

PwC offers different model risk management technology platforms to help your organization manage model inventory, reporting, and all aspects of the model lifecycle across all stakeholder groups. We offer a PwC-developed solution as well as solutions that are based on industry leading GRC platforms.

AI/ML model risk governance
Responsible AI
Internal audit support

AI/ML model risk governance

With the financial services industry rapidly embracing analytical solutions powered by artificial intelligence / machine learning models (AI/ML), managing unique risks associated with the use of these algorithms becomes increasingly important. In helping our clients design effective AI/ML model risk management programs, we draw on our hands-on experience in designing and implementing machine learning models along with our deep understanding of effective and sustainable model risk management practices that meet regulatory requirements.

Validation of in-house developed and vendor-based AI/ML models require specialized knowledge and expertise. Our validation specialists combine deep understanding of AI/ML modeling techniques with broad subject matter expertise required to assess unique risks associated with each model use case and produce a regulatory compliant model validation.

Responsible AI

Broad adoption of AI/ML combines great promise with elevated scrutiny from customers and regulators. Responsible AI Toolkit is a suite of customizable frameworks, tools and processes designed to help you harness the power of AI in an ethical and responsible manner - from strategy through execution. With the Responsible AI toolkit, we’ll tailor our solutions to address your organisation’s unique business requirements and AI maturity.

Our experience with MRM practices across a multitude of financial service firms and our deep understanding of regulatory expectations makes us a leading provider of model governance services and gap assessments relative to the industry best practices.

Competencies, general process / testing plan, and case studies.

Our offerings in the governance space include:

Gap assessment relative to the industry best practices and regulatory expectations;
Recommendation for enhancement of MRM programs to address emergent risk, including growing use of AI/ML models
Redesign of MRM programs in response to regulatory criticism or lack of effectiveness of certain elements.

Internal audit support

Our team has been at the forefront of bringing model risk management expertise to our internal audit clients. Our internal audit service offering includes assistance in assessment of effectiveness of the clients’ independent model risk programs, including model identification, inventorying, use, and change management. Special attention is always paid to the model development and model validation activities, as those are the areas of continued focus by the regulators.

Our broader team has expertise across all areas of economic and financial modeling, including credit risk models, interest rate risk and liquidity, asset and liability, market risk, BSA/AML, and financial reporting models. This expertise allows us to effectively and efficiently support audits across a variety of business areas with significant reliance on models.

In addition to conducting model-specific audits, our service offerings to internal audit teams in the model risk management space have included assistance in the development or enhancement of MRM internal audit programs, including methodology and templates, as well assistance in assessment of the design of the bank’s independent model risk management program, including MRM policies and procedures.

Competencies, general process / testing plan and case studies.

Our offerings in the internal audit space include:

Audit of model risk management program design and operational effectiveness;
Model-specific audits;
Designing and enhancing MRM internal audit programs

Model development & acquisition

The model lifecycle begins with the definition of business and technical requirements, evaluation of existing models that may satisfy the requirements, and evaluation of options for internal development vs. vendor model acquisition.

For internally developed models, institutions typically follow formal development and testing processes and standards. The larger is the pool of similar models, the greater is the opportunity to automate model development, testing and documentation processes.

For external acquired models, institutions typically follow a robust due diligence process to ensure that the models developed in industry data will meet their needs and perform well on the internal products, customers and transactions.

How can we help?

Model development/acquisition planning

PwC model governance specialists can help you develop or enhance your existing planning and acquisition processes, standards, and controls as part of the model risk management framework, as well as the third party risk management framework (for external models).
PwC offers model risk management technology platforms to manage the full model lifecycle, model inventory and model risk reporting.

Model development

PwC offers model development services for a broad range of models, including the following areas:

Credit risk
Market risk
Revenue, expenses and balance sheet forecasting

Our products and accelerators can help you automate model development and testing activities, as well as production of model documentation, yielding greater efficiency as well as quality and consistency.

Model development services

Credit risk modeling
Market risk modeling
Revenue, expenses & balance sheet modeling

Credit risk modeling

We assist our clients with development of a wide range of credit risk models, including but not limited to:

Consumer and commercial credit models, such as CCAR/DFAST stress testing models, credit scorecards, credit loss forecasting models, and models used to support Basel capital estimates.
Financial reporting models, such as ASC 310-20 (formerly FAS 91) amortization, ASC 450 (formerly FAS 5) reserves, CECL, IFRS 9, and ASC 310-30 (formerly SOP 03-3).

Our experiences and expertise cover a broad range of modeling approaches, from top-down time series OLS regressions, to competing risk hazard rate models, dynamic transition matrix approaches driven by statistical regressions, as well as advanced machine learning approaches.

Market risk modeling

We assist our clients with development of a wide range of market risk models, including but not limited to:

Financial instrument attribution models
VaR time series development for legacy assets
Liquidity surface development
Mortgage loan prepayment and valuation models
Customized Ongoing Performance Assessment (OPA) reports
What-if tools for regulatory and economic capital models
Dynamic risk limits models for ex-ante and ex-post Pre-settlement Exposure (PSE) impacts
CVA/PSE divergence models
Concentration risk optimization models

Revenue, expenses & balance sheet modeling

In the revenue and balance sheet modeling practice, we assist our clients develop robust quantitative models to support financial planning and capital planning processes.

Our models capture the changes in PPNR (Pre-provision Net Revenue) and balance sheet of the businesses during stress scenarios such as the financial crisis and COVID impacts, as well during baseline scenarios

We have successfully developed models that cover a wide range of business and types of activities performed by banks including:

Retail and relationship
- Retail lending
- Wealth management
- Investment management
Investment banking
- Underwriting and advisory
- Sales and trading
- Commercial lending
- Transaction banking

We follow a rigorous model development process with the following key principles:

Priority for intuition of the model over statistical fit
Basis of the forecasts meet the expectations of the business owners
Statistical soundness
Risk capture

Model implementation

Model implementation refers to the hand-off of the model from the “sand-box” development environment into the production environment, where the model is deployed to produce outputs consumed by the business or downstream applications.

Before the model goes “live” in the production environment, comprehensive and thorough implementation testing is needed to ensure the model is implemented as intended and is consistent with the documented specifications. Additional user acceptance testing (UAT) may be required to confirm that the model outputs meets the business use requirements. Regulators and the MRM function expect robust documentation around the design and outcomes of implementation testing.

How can we help?

Model implementation controls

PwC model governance specialists can help you develop or enhance your existing model implementation processes and controls as part of the model risk management framework.
PwC offers model risk management technology platforms to manage the full model lifecycle, model inventory, and model risk reporting.

Model implementation testing

PwC offers model implementation testing as part of the model development services.

Model validation

Independent model validation is a cornerstone of effective model risk management. To ensure effective challenge to the model developers, validators must be independent and possess a combination of technical skills as well as the relevant domain business expertise.

How can we help?

With over 20 years of model validation service history, PwC can help you with ongoing or surge needs for validating any type of model. We also offer a range of products and accelerators that can help you automate your internal validation testing to improve efficiency and quality.

PwC independent model validation services

Credit risk models
Market risk models
PPNR models
AML, sanctions screening and fraud models
Cyber security models
Artificial intelligence / Machine learning models

Model validation automation

Our products and accelerators can help you automate model validation testing activities, as well as production of model validation and periodic review reports, yielding greater efficiency as well as quality and consistency.

Model validation services

Credit risk model validation
Market risk model validation
Revenue & balance sheet model validation
AML, sanctions screening, fraud model validation

Credit risk model validation

We provide leading services focused on assistance with independent model validations for a wide range of credit risk models, including:

Consumer and commercial credit models, including CCAR/DFAST stress testing models, credit scorecards, credit loss forecasting models, and models used to support Basel capital estimates
Financial reporting models, including ASC 310-20 (formerly FAS 91) amortization, ASC 450 (formerly FAS 5) reserves, CECL, IFRS 9, and ASC 310-30 (formerly SOP 03-3)

Market risk model validation

We provide leading services focused on assistance with independent model validations for a wide range of credit and behavioural risk models, including:

Agency and non-agency prepayment forecasting and valuation models
Whole loan and correspondent channel credit and prepayment valuation models
CMO and non-agency loan-level credit curve forecasting and valuation models
Valuation for non-agency rated structured products models
Financial products valuation
FRTB driven model assessment for illiquid legacy structured products assets
Liquidity valuation models
WWR valuation models

Revenue & balance sheet model validation

We incorporate industry leading practices in our model validation procedures and align our solution to with the bank’s internally developed standards and templates

We provide:

Model specific testing plan based on model risk assessment
Consistency through uniformity in methodologies and processes used across all model validators
Hierarchy of responses delivered through standardized template
Enhanced library of omdel risks and validation testing procedures

AML, sanctions screening, fraud model validation

Our dedicated team of financial crimes specialists offers model validation services in the following areas:

Anti-money laundering: transaction monitoring and customer risk rating models
Sanctions: sanctions screening and watch list filtering models
Fraud monitoring: entity monitoring, session monitoring, payment transaction monitoring models

Our model validation testing procedures are designed to identify risks and limitations associated with your use of these types of models and to identify recommendations for addressing findings for management’s consideration, which can be developed into an actionable strategic and tactical roadmap.

Ongoing monitoring

Periodic monitoring of active models is a key part of the model lifecycle that helps ensure the models are performing as intended and continue to meet the business use requirements.

Effective performance monitoring combines alignment of monitoring metrics, frequency of monitoring, and monitoring thresholds with the model type, business use, and inherent risk tier.

Robust risk monitoring can help identify early warning signs that the model is no longer accurate and needs to be redeveloped, recalibrated or otherwise adjusted, due, for example, to the changes in the economic/business environment or customer behavior. Comprehensive and timely model monitoring is especially important for AI/ML models that are often complex, rely on massive quantities of data, and may be prone to bias and rapid performance deterioration.

How can we help?

PwC model governance specialists can help you develop or enhance your existing model monitoring processes, standards and controls as part of the model risk management framework.
PwC offers model risk management Technology platforms to manage the full model lifecycle, model inventory, and model risk reporting.
We can help design or enhance your ongoing monitoring process for a broad range of models, to ensure the monitoring metrics and frequency are aligned with model use and regulatory expectations.
Our products and accelerators can help you automate ongoing monitoring activities, including: monitoring report production to increase efficiency, quality and consistency of the monitoring process.

Risk analytics

Overview

PwC risk analytics professionals can help you with a variety of ad hoc data analysis using a broad range of industry tools, as well as with the development of analytical dashboards and reporting systems.

Details of services

Credit risk analysis
Fair lending compliance

Credit risk analysis

Our credit professionals possess many years of experience in industry as bankers, credit underwriters, credit risk and loan loss provision management professionals, regulators, and auditors. We have experience with a broad range of financial institutions, including both foreign based and US based large global banks, and US regional banks, specifically in the area of credit policies, underwriting, risk rating systems, credit risk measurement, and credit administration procedures.

The PwC team has broad exposure to industry leading credit analytics, policies and related processes. We have helped clients create new credit policies, revise current policies and designed risk rating methodologies and templates utilizing expert and statistical design processes. We have trained bank staff on US regulatory expectations and have designed job aids.

We leverage our world-wide network of professionals to help clients solve a variety of complex credit risk and regulatory challenges. As global auditors, we have built strong relationships with several leading institutions and have a deep knowledge of credit risk measurement and credit risk operations.

Because we have worked with a broad range of financial institutions, we are able to bring industry leading practices and tailor them to your specific business, culture and capabilities.

The true value in our team is the thoughtful approach we bring to solving complex credit risk problems. Our experienced team offers relevant and impactful leading practices which help clients navigate the complex business and regulatory environment.

How risky is your loan portfolio? Three questions for lenders to ask

Fair lending compliance

With our blend of regulatory experience, industry knowledge, and state-of-the-industry technologies, we are well-positioned to assist you with your fair lending compliance risk management needs. Specifically:

Fair lending risk and control assessments

Identification of fair lending risks requiring attention – either enterprise-wide or for specific business units – through design and performance of “mock” fair lending exams
Benchmarking of fair lending compliance programs and procedures to regulatory expectations and industry leading practices
Provide recommendations to improve compliance programs and procedures
Advice on the design and operational efficiency of fair lending control processes
Independent evaluation of the design and implementation of fair lending regression models and monitoring systems

Fair lending testing: Using widely accepted regulatory and industry proxy methodologies for race / ethnicity and gender to proxy code data and identify protected class borrowers; Perform customized, statistically-based testing on your entire loan application or origination population to identify potential fair lending risks associated with the following lending outcomes:

Redlining
Credit decisions
Loan Pricing (APR, Rate Spread, Dealer Mark-up)
Steering
Servicing

Fair Lending Monitoring Programs

Perform independent testing of the accuracy of certain HMDA fields (e.g., APR rate spreads)
Design and build statistically-based fair lending monitoring systems for on-going fair lending compliance risk management
Provide advice on the design of fair lending corrective action policies
Assist with manual file reviews

Fair Lending Model Validation

Assess model’s conceptual soundness and consistency with regulatory requirements and industry approaches
Evaluate model’s technical soundness and statistical validity
Review model implementation accuracy and data appropriateness

Fair lending due diligence: For a proposed merger or acquisition, we assess the acquirees potential fair lending risks as well as the quality of its fair lending compliance risk management system; and

Fair lending litigation support: Our experience in the statistical analysis of large databases, and our understanding of the fair lending risks inherent to all stages of the lending process, position us well to assist inside and outside counsel with their response to government investigations and civil litigation.

Our commitment to client service means that PwC brings not only the right skills and resources for your specific needs, but also resources that have the right combination of consumer lending knowledge, experience with fair lending regulatory expectations and examination procedures, and strong statistical modeling and analytics capabilities. We believe experience in all three areas is critically important in providing value to our clients, and in building long-term trusted business relationships.

Model Risk Management Technology Solutions

Depending on your needs and existing technological infrastructure, PwC can provide you with MRM Technology Solutions that fit your needs. Model Edge is a PwC product offering a cost-effective cloud-based governance solution. If you would rather build up an MRM solution based on your existing Governance Risk and Compliance platform, PwC offers custom developed solutions based on the leading industry GRC platforms including Archer, MetricStream, ServiceNow, and others.

Depending on your institution’s needs and preferences, we can offer you a PwC-developed product or a solution developed on top of your existing Governance Risk and Compliance (GRC) platform from another vendor:

Model Edge. A PwC Product

Model Edge is a PwC in-house platform developed to customize automation solutions for model development, model validation, and model documentation. The automation toolkit significantly boosts the work efficiency, and brings in the potential to re-develop models with limited efforts.

MRM Technology Solutions

PwC’s Model Risk Management Technology Solutions customize multiple leading GRC platforms to offer effective MRM technology frameworks for maintaining model inventory, manage processes and workflows at each stage of model lifecycle, and provide model risk reporting capabilities.

Case studies

1 result

29/05/20

How insurance claims estimators can use AI to increase efficiency, reduce cycle time and improve customer experience

Using “explainable AI” and image recognition technology to scale insurance claims estimation process

Contact us

Jason Dulnev

Principal, Risk Analytics, PwC US

Hide

Required fields are marked with an asterisk(*)

First Name*

Last Name*

Company Name*

Job Title*

Job Function*

Email address*

Location*

By submitting your email address, you acknowledge that you have read the Privacy Statement and that you consent to our processing data in accordance with the Privacy Statement (including international transfers). If you change your mind at any time about wishing to receive the information from us, you can send us an email message using the Contact Us page.

Audit and Assurance services Consulting Tax services Newsroom Alumni US offices Contact us

© 2017 - 2025 PwC. All rights reserved. PwC refers to the PwC network and/or one or more of its member firms, each of which is a separate legal entity. Please see www.pwc.com/structure for further details.

Overview

The financial risk analytics and modeling lifecycle

Model governance

How can we help?

AI/ML model risk governance

Model Risk Management Technology Solutions

Case studies

How insurance claims estimators can use AI to increase efficiency, reduce cycle time and improve customer experience

Related content

Financial Services Risk and Regulatory

Contact us

Thank you for your interest in PwC