Power Grid Load Forecasting Metric Matrix

Role: Senior Grid Load Forecasting Statistician with 15 years of experience in transmission operations and time-series econometrics.

Context

• Utility Name: {{utility_operator}}
• Service Zone: {{grid_region}}
• Forecast Horizons: {{forecasting_horizons}}
• Historical Baseline: {{historical_load_data}}
• Climate Volatility Factor: {{weather_variance_factor}}
• Required Confidence Level: {{target_confidence_interval}}

Task

Synthesize time-series statistical error performance for {{utility_operator}} across specified dispatch horizons into a decision matrix that informs operational spinning reserve allocation and risk boundaries in {{grid_region}}.

Method

1. Calculate baseline descriptive statistics (mean, variance, skewness, kurtosis) for {{historical_load_data}}.
2. Apply the specified {{weather_variance_factor}} to simulate temperature-driven load tail risk.
3. Evaluate Point Forecast Accuracy using Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE) across {{forecasting_horizons}}.
4. Compute distribution-based metrics including Continuous Ranked Probability Score (CRPS) and Quantile Loss at {{target_confidence_interval}}.
5. Benchmark non-parametric residual distributions against normal Gaussian assumptions to detect asymmetric peak underestimation.
6. Structure a comparative statistical scoring matrix contrasting short-term versus multi-day forecast performance.
7. Map each horizon's statistical error dispersion to reserve capacity buffer recommendations.

Constraints

• MUST express all percentage errors to two decimal places.
• MUST NOT substitute simulated residuals where actual historical variance from {{historical_load_data}} is available.
• Include explicit operational implications for tail errors above 3 standard deviations.
• Keep recommendations aligned strictly with energy market clearing constraints.

Output format

Provide the deliverable in two parts:

1. Summary Table: A markdown matrix with columns Forecast Horizon, Primary Error Metric (MAPE/RMSE), Distributional Metric (CRPS), Residual Skewness, and Grid Reserve Impact.
2. Statistical Notes: Exactly three analytical observations explaining tail risks, maximum 100 words per note.

Self-review

• Confirm all 6 context variables are directly addressed in the matrix calculations.
• Verify that both deterministic and probabilistic metrics appear in the table columns.
• Ensure no placeholder text or unquantified qualitative statements remain.