[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F89V1RHHNDXCSS83NYBZ55":3,"skill-same:01M7F89V1RHHNDXCSS83NYBZ55":65},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":14,"repositoryHtmlUrl":48,"repositoryHomepageUrl":48,"ownerHtmlUrl":49,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":50,"install":51,"resources":64},"01M7F89V1RHHNDXCSS83NYBZ55","managing-credit-risk-models",null,"Evaluates and monitors credit risk models (PD, LGD, EAD) with calibration and discrimination metrics. Use when validating credit models, assessing model performance, or calibrating default models.","Mr-Nobody-Anonymous\u002FAll-skills","Mr-Nobody-Anonymous","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F184253119?v=4","01M7AEXFJHKTYQW5R0ZCQQTWFV","anonymous","01M7C00M3GQR4QKP9GZA42GFAK",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Ffinance\u002Fmanaging-credit-risk-models\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: managing-credit-risk-models\nlanguage: en\ndescription: Evaluates and monitors credit risk models (PD, LGD, EAD) with calibration and discrimination metrics. Use when validating credit models, assessing model performance, or calibrating default models.\ntags:\n  - management\n  - risk-management\n  - risk\n  - credit\nmetadata:\n  author: casemark\n  practice_areas:\n    - Risk Management\n    - Enterprise Risk\n    - Market Risk\n  document_types:\n    - Management Report\n  skill_modes:\n    - Management\n    - Coordination\nsource: \"https:\u002F\u002Fgithub.com\u002FCaseMark\u002Fskills\"\nsource_repository: \"CaseMark\u002Fskills\"\nsource_path: \"skills\u002Ffinance\u002Fmanaging-credit-risk-models\u002FSKILL.md\"\nlicense: \"Apache-2.0\"\nimported_at: \"2026-09-20\"\n---\n# Managing Credit Risk Models\n\n## When To Use\n\n- Periodic validation of Probability of Default (PD), Loss Given Default (LGD), or Exposure at Default (EAD) models\n- Annual or triggered model performance reviews required by internal model governance or regulatory mandate (e.g., SR 11-7, CRD\u002FCRR, IFRS 9 ECL frameworks) [VERIFY jurisdiction-specific regulatory requirements]\n- Recalibration following material portfolio shifts, macroeconomic regime changes, or post-merger data integration\n- Comparing challenger models against incumbent production models before promotion\n- Preparing model risk management reports for MRC\u002Fboard-level review\n\n## Inputs To Gather\n\n- **Development sample and out-of-time\u002Fout-of-sample validation data** — confirm vintage coverage, default definition consistency, and exclusion criteria\n- **Model documentation** — original methodology paper, variable selection rationale, and any prior validation findings\n- **Current production scorecards or parameter estimates** — PD term structures, LGD cure\u002Fworkout assumptions, EAD CCF tables\n- **Realized outcome data** — observed default flags, recovery cashflows, drawn\u002Fundrawn balances at default\n- **Portfolio segmentation** — rating grades, facility types, collateral categories, geographic or industry cuts\n- **Regulatory and policy thresholds** — minimum discrimination (e.g., Gini > 0.40), calibration tolerance bands, override rate caps [VERIFY institution-specific thresholds]\n\n## Workflow\n\n1. **Scope the validation exercise**\n   - Identify which model components are in scope (PD only, full PD\u002FLGD\u002FEAD suite, segment-level vs. portfolio-level)\n   - Confirm the observation window and default\u002Floss outcome definitions match the model's design assumptions\n   - Document any data limitations or exclusions upfront\n\n2. **Assess discrimination performance**\n   - Compute Gini coefficient (Accuracy Ratio), AUC-ROC, and Kolmogorov-Smirnov statistic on the validation sample\n   - Generate CAP (Cumulative Accuracy Profile) and ROC curves\n   - Segment discrimination by key risk drivers (vintage, industry, geography) to detect pockets of weakness\n   - Compare current-period metrics against development-sample benchmarks and prior validation results\n\n3. **Evaluate calibration accuracy**\n   - Run Binomial test, Hosmer-Lemeshow test, or traffic-light approach (Basel) across PD buckets\n   - For LGD: compare predicted vs. realized loss severity by collateral type and workout path\n   - For EAD: compare predicted CCFs against observed utilization at default\n   - Assess calibration across economic cycles — flag if model was calibrated to benign conditions and current environment is stressed [VERIFY whether TTC vs. PIT calibration applies]\n\n4. **Evaluate stability and concentration**\n   - Population Stability Index (PSI) on score distributions between development and recent periods\n   - Characteristic Stability Index (CSI) on key input variables\n   - Herfindahl index or grade-concentration analysis to detect rating migration clustering\n   - Flag PSI > 0.25 or CSI > 0.25 as material shifts requiring deeper investigation [VERIFY institution-specific PSI thresholds]\n\n5. **Stress-test and sensitivity analysis**\n   - Perturb key macro drivers (GDP, unemployment, HPI) and assess PD\u002FLGD migration under stressed scenarios\n   - Identify variables with outsized sensitivity — single-variable stress contributions exceeding a defined threshold\n   - Cross-check stressed outputs against institution's CCAR\u002FDFAST or ICAAP submissions if applicable [VERIFY regulatory stress testing framework]\n\n6. **Document findings and recommend actions**\n   - Classify findings by severity: Tier 1 (material, requires remediation before next use), Tier 2 (significant, remediation within defined timeline), Tier 3 (minor, monitor)\n   - Provide specific recalibration or redevelopment recommendations with target timelines\n   - Draft executive summary for Model Risk Committee or board reporting\n\n## Output\n\n- **Model Validation Report** containing:\n  - Executive summary with overall model rating (e.g., Satisfactory \u002F Needs Improvement \u002F Unsatisfactory)\n  - Discrimination metrics table (Gini, AUC, KS) with trend comparison across validation periods\n  - Calibration test results by grade, segment, and time horizon\n  - Stability analysis with PSI\u002FCSI tables and heatmaps\n  - Findings register with severity tier, description, and remediation action\u002Fowner\u002Fdeadline\n  - Appendices: data quality notes, exclusion log, detailed statistical output\n\n## Quality Checks\n\n- Confirm default and loss definitions used in validation match the model's training definitions exactly — misalignment here invalidates all downstream metrics\n- Verify that validation data has no lookahead bias (outcomes must post-date the score assignment)\n- Cross-check sample sizes per rating grade — bins with fewer than 30 defaults produce unreliable calibration test results\n- Ensure discrimination and calibration metrics are computed on the same population; filtered vs. unfiltered samples can yield contradictory conclusions\n- Validate that any override or judgmental adjustment rates are reported separately and not mixed into statistical performance metrics\n- Confirm findings are mapped to specific model components — avoid blanket \"model is weak\" conclusions without identifying which parameter (PD, LGD, or EAD) and which segment drives the issue\n","# Managing Credit Risk Models\n\n## When To Use\n\n- Periodic validation of Probability of Default (PD), Loss Given Default (LGD), or Exposure at Default (EAD) models\n- Annual or triggered model performance reviews required by internal model governance or regulatory mandate (e.g., SR 11-7, CRD\u002FCRR, IFRS 9 ECL frameworks) [VERIFY jurisdiction-specific regulatory requirements]\n- Recalibration following material portfolio shifts, macroeconomic regime changes, or post-merger data integration\n- Comparing challenger models against incumbent production models before promotion\n- Preparing model risk management reports for MRC\u002Fboard-level review\n\n## Inputs To Gather\n\n- **Development sample and out-of-time\u002Fout-of-sample validation data** — confirm vintage coverage, default definition consistency, and exclusion criteria\n- **Model documentation** — original methodology paper, variable selection rationale, and any prior validation findings\n- **Current production scorecards or parameter estimates** — PD term structures, LGD cure\u002Fworkout assumptions, EAD CCF tables\n- **Realized outcome data** — observed default flags, recovery cashflows, drawn\u002Fundrawn balances at default\n- **Portfolio segmentation** — rating grades, facility types, collateral categories, geographic or industry cuts\n- **Regulatory and policy thresholds** — minimum discrimination (e.g., Gini > 0.40), calibration tolerance bands, override rate caps [VERIFY institution-specific thresholds]\n\n## Workflow\n\n1. **Scope the validation exercise**\n   - Identify which model components are in scope (PD only, full PD\u002FLGD\u002FEAD suite, segment-level vs. portfolio-level)\n   - Confirm the observation window and default\u002Floss outcome definitions match the model's design assumptions\n   - Document any data limitations or exclusions upfront\n\n2. **Assess discrimination performance**\n   - Compute Gini coefficient (Accuracy Ratio), AUC-ROC, and Kolmogorov-Smirnov statistic on the validation sample\n   - Generate CAP (Cumulative Accuracy Profile) and ROC curves\n   - Segment discrimination by key risk drivers (vintage, industry, geography) to detect pockets of weakness\n   - Compare current-period metrics against development-sample benchmarks and prior validation results\n\n3. **Evaluate calibration accuracy**\n   - Run Binomial test, Hosmer-Lemeshow test, or traffic-light approach (Basel) across PD buckets\n   - For LGD: compare predicted vs. realized loss severity by collateral type and workout path\n   - For EAD: compare predicted CCFs against observed utilization at default\n   - Assess calibration across economic cycles — flag if model was calibrated to benign conditions and current environment is stressed [VERIFY whether TTC vs. PIT calibration applies]\n\n4. **Evaluate stability and concentration**\n   - Population Stability Index (PSI) on score distributions between development and recent periods\n   - Characteristic Stability Index (CSI) on key input variables\n   - Herfindahl index or grade-concentration analysis to detect rating migration clustering\n   - Flag PSI > 0.25 or CSI > 0.25 as material shifts requiring deeper investigation [VERIFY institution-specific PSI thresholds]\n\n5. **Stress-test and sensitivity analysis**\n   - Perturb key macro drivers (GDP, unemployment, HPI) and assess PD\u002FLGD migration under stressed scenarios\n   - Identify variables with outsized sensitivity — single-variable stress contributions exceeding a defined threshold\n   - Cross-check stressed outputs against institution's CCAR\u002FDFAST or ICAAP submissions if applicable [VERIFY regulatory stress testing framework]\n\n6. **Document findings and recommend actions**\n   - Classify findings by severity: Tier 1 (material, requires remediation before next use), Tier 2 (significant, remediation within defined timeline), Tier 3 (minor, monitor)\n   - Provide specific recalibration or redevelopment recommendations with target timelines\n   - Draft executive summary for Model Risk Committee or board reporting\n\n## Output\n\n- **Model Validation Report** containing:\n  - Executive summary with overall model rating (e.g., Satisfactory \u002F Needs Improvement \u002F Unsatisfactory)\n  - Discrimination metrics table (Gini, AUC, KS) with trend comparison across validation periods\n  - Calibration test results by grade, segment, and time horizon\n  - Stability analysis with PSI\u002FCSI tables and heatmaps\n  - Findings register with severity tier, description, and remediation action\u002Fowner\u002Fdeadline\n  - Appendices: data quality notes, exclusion log, detailed statistical output\n\n## Quality Checks\n\n- Confirm default and loss definitions used in validation match the model's training definitions exactly — misalignment here invalidates all downstream metrics\n- Verify that validation data has no lookahead bias (outcomes must post-date the score assignment)\n- Cross-check sample sizes per rating grade — bins with fewer than 30 defaults produce unreliable calibration test results\n- Ensure discrimination and calibration metrics are computed on the same population; filtered vs. unfiltered samples can yield contradictory conclusions\n- Validate that any override or judgmental adjustment rates are reported separately and not mixed into statistical performance metrics\n- Confirm findings are mapped to specific model components — avoid blanket \"model is weak\" conclusions without identifying which parameter (PD, LGD, or EAD) and which segment drives the issue\n",{"name":5,"tags":24,"source":29,"license":15,"language":30,"metadata":31,"description":7,"imported_at":42,"source_path":43,"source_repository":44},[25,26,27,28],"management","risk-management","risk","credit","https:\u002F\u002Fgithub.com\u002FCaseMark\u002Fskills","en",{"author":32,"skill_modes":33,"document_types":36,"practice_areas":38},"casemark",[34,35],"Management","Coordination",[37],"Management Report",[39,40,41],"Risk Management","Enterprise Risk","Market Risk","2026-09-20","skills\u002Ffinance\u002Fmanaging-credit-risk-models\u002FSKILL.md","CaseMark\u002Fskills","main","2d1750717747097b602acf30bd09a7cbd4b4c6ea","awesome_skills\u002Ffinance\u002Fmanaging-credit-risk-models","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous","2026-11-09 11:45:36",{"source":8,"skillName":5,"commands":52,"prompt":62,"zipAvailable":63},[53,56,59],{"tool":54,"command":55},"npx","npx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill managing-credit-risk-models",{"tool":57,"command":58},"bunx","bunx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill managing-credit-risk-models",{"tool":60,"command":61},"pnpm","pnpm dlx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill managing-credit-risk-models","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Ffinance\u002Fmanaging-credit-risk-models\u002FSKILL.md",true,[],[66,74,81,88,95,102],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M7FFFZPC2025JQQHPHDX154J","dei-assessment","Assess diversity, equity, and inclusion with metrics. TRIGGERS - Use when user needs help with dei-assessment related tasks.","MIT","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-assessment\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7FFFWN8H4WA7A6Y398QKB6J","dei-action-plan","Create DEI action plans with goals and initiatives. TRIGGERS - Use when user needs help with dei-action-plan related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-action-plan\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M7FFFSHTM0QZRVZ2NRJTJ0PY","culture-fit-assessment","Design culture add assessment processes. TRIGGERS - Use when user needs help with culture-fit-assessment related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fculture-fit-assessment\u002FSKILL.md",[],[],{"skillNo":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M7FFFMA12RKEQA38KHJSXXPS","cowork-to-code-bridge","Use an already-installed, independently verified cowork-to-code bridge to run narrowly approved actions on the user's own macOS, Linux, or WSL2 machine through a local file queue.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcowork-to-code-bridge\u002FSKILL.md",[],[],{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":19},"01M7FFFH1F6M8YRGW5V487GH1K","counter-offer-response","Draft counter-offer responses for retention. TRIGGERS - Use when user needs help with counter-offer-response related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcounter-offer-response\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":70,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":19},"01M7FFFDGEHV8Y7J9N40HZXW52","continuous-feedback-system","Design continuous feedback systems. TRIGGERS - Use when user needs help with continuous-feedback-system related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcontinuous-feedback-system\u002FSKILL.md",[],[]]