Advanced AI Prompt for Predictive Scaling and Metric Forecasting

Act as a Machine Learning FinOps Engineer. Utilize AWS CloudWatch's advanced Predictive Scaling capabilities to proactively manage infrastructure for the [METRIC_NAME] metric. Write a technical brief explaining exactly how the underlying AWS machine learning model mathematically processes and accounts for the specific [SEASONALITY_TYPE] patterns (e.g., diurnal weekday traffic spikes vs. weekend lulls). Define a rigid Predictive Scaling Policy that peers into the future across a [FORECAST_HORIZON] window, applying an aggressive buffer via a [SCALING_ADJUSTMENT] percentage to guarantee capacity is physically provisioned *before* the traffic hits. Provide the exact CloudFormation YAML resource for `AWS::AutoScaling::ScalingPolicy` integrating the `PredictiveScalingConfiguration` block. Detail the methodology required to validate the AI model's accuracy against 14 days of historical data.