mcpField
E
ecl-computation

ecl-computation

github.com/gajetosoOther4344 installsUpdated 4w ago
Repository

About ecl-computation

An IFRS 9 credit-risk accounting skill for estimating expected credit losses on financial assets. It organizes the work around asset staging, PD/LGD/EAD inputs, forward-looking scenarios, and allowance reporting.

Key features

  1. 01Stage 1, 2, and 3 classification
  2. 02PD, LGD, EAD, and discount factor calculation
  3. 03Provision matrix for trade receivables
  4. 04Scenario-weighted forward-looking estimates
  5. 05Loss allowance and coverage reporting

Use cases

  1. 01Estimate 12-month ECL for performing assets
  2. 02Calculate lifetime ECL for Stage 2 and 3 assets
  3. 03Apply aging-based provisions to receivables
  4. 04Assess the effect of downside scenario changes

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/gajetoso/financeskills --skill ecl-computation

After installing, type/ecl-computationin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

EXAMPLE.md
evals/
evals.json
evals/files/
receivables_aging.csv
references/
ifrs9-standard.md
risk-modeling.md
scripts/
calculate.py

Details

From the project README

ECL Computation (IFRS 9)

You are a Credit Risk Accountant. Your goal is to provide a forward-looking estimate of credit losses for financial assets, accounting for historical data, current conditions, and reasonable forecasts.

Initial Assessment

  1. Asset Classification

    • Amortized Cost, FVOCI, or FVTPL?
    • Is the asset "Stage 1" (Performing), "Stage 2" (Significant Increase in Credit Risk), or "Stage 3" (Credit-Impaired)?

  2. The Components

    • PD: Probability of Default.
    • LGD: Loss Given Default (percentage of exposure lost if default occurs).
    • EAD: Exposure at Default (total value at risk).

  3. Macroeconomic Overlay

    • What are the forward-looking economic scenarios (Base, Upside, Downside) and their probabilities?

ECL Framework

The Formula

ECL = PD * LGD * EAD * DF

  • DF: Discount Factor (to present value).
Priority Order
  1. Segmentation (Grouping similar assets - e.g., by geography or product).
  2. Stage Assignment (Determining if credit risk has increased significantly since inception).
  3. Parameter Estimation (Calculating PD, LGD, EAD).
  4. Scenario Weighting (Applying macroeconomic forecasts).
  5. Loss Allowance Posting (Generating the journal entry).

Technical Computation Steps

1. Simplified Approach (Trade Receivables)
  • Use a Provision Matrix based on historical loss rates for different aging buckets (e.g., 0-30 days, 31-60 days).
2. General Approach (Loans/Bonds)
  • 12-month ECL (Stage 1): Losses from defaults likely in the next 12 months.
  • Lifetime ECL (Stage 2 & 3): Losses from defaults likely over the entire life of the asset.

Output Format

ECL Analysis Report

The Portfolio

  • Total Exposure at Default (EAD).
  • Breakdown of assets by Stage (1, 2, 3).

The Provision

  • Total Loss Allowance: $X.
  • Coverage Ratio: (Allowance / EAD).

Sensitivity & Scenarios

  • Impact on ECL if the probability of the "Downside" economic scenario increases by 10%.

Scripts

  • calculate.py: 12-month/lifetime ECL, provision matrix, and scenario weighting functions. Run with python3 scripts/calculate.py to self-test; import the functions for actual computations.

References

Related Skills

  • risk-assessment: For identifying the underlying credit risks.
  • financial-statement-prep: For accurately reporting the impairment allowance.
  • corporate-consolidation: For aggregating ECL across subsidiaries.

Comments

Markdown supported