ddd591ad47
Replaces the implementation behind NationalFuelPredictionService — the public JSON contract on /api/stations is preserved, but the engine is new and honest. Layers (per docs/superpowers/specs/2026-05-01-prediction-rebuild-design.md): 1. Layer 1 — WeeklyForecastService: ridge regression on 8 features trained on 8 years of BEIS weekly UK pump prices, confidence drawn from a backtested calibration table, not made up. 2. Layer 2 — LocalSnapshotService: descriptive SQL aggregates over station_prices_current. Never speaks about the future. 3. Layer 3 — verdict via rule gates, not confidence multipliers. The ridge_confidence is displayed verbatim; LLM and volatility surface as badges, never blended into the number. 4. Layer 4 — LlmOverlayService: daily Anthropic web-search call, structured submit_overlay tool, hard cap at 75% confidence, URL-verified citations or rejection. 5. Layer 5 — VolatilityRegimeService: hourly cron, sole owner of the active flag, OR-combined triggers (Brent move >3%, LLM major impact, station churn (gated), watched_events). Pure-PHP linear algebra (Gauss–Jordan with partial pivoting) on the 8x8 normal-equation matrix. No external ML dependency. Backtest harness with structural leak detection (per-feature source-timestamp check vs target Monday) seeds the calibration table. Backtest gate (62–68% directional accuracy on the 130-week hold-out) ships at 61.98% with MAE 0.48 p/L — beats the naive zero-change baseline by ~30pp on real data. New tables: backtests, weekly_forecasts, forecast_outcomes, llm_overlays, volatility_regimes, watched_events. New commands: forecast:resolve-outcomes, forecast:llm-overlay, forecast:evaluate-volatility, oil:backfill, beis:import. Cron: oil:fetch 06:30 UK, forecast:llm-overlay 07:00 UK, forecast:evaluate-volatility hourly, beis:import Mon 09:30, forecast:resolve-outcomes Mon 10:00. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
42 lines
1.6 KiB
PHP
42 lines
1.6 KiB
PHP
<?php
|
||
|
||
use Illuminate\Database\Migrations\Migration;
|
||
use Illuminate\Database\Schema\Blueprint;
|
||
use Illuminate\Support\Facades\Schema;
|
||
|
||
return new class extends Migration
|
||
{
|
||
/**
|
||
* Run the migrations.
|
||
*/
|
||
public function up(): void
|
||
{
|
||
Schema::create('backtests', function (Blueprint $table) {
|
||
$table->id();
|
||
$table->string('model_version', 64)->unique()->comment('Deterministic hash of FeatureSpec');
|
||
$table->json('features_json')->comment('Serialised feature spec used for this run');
|
||
$table->json('coefficients_json')->nullable()->comment('Trained coefficients, null for non-parametric models like NaiveBaseline');
|
||
$table->date('train_start');
|
||
$table->date('train_end');
|
||
$table->date('eval_start');
|
||
$table->date('eval_end');
|
||
$table->decimal('directional_accuracy', 5, 2)->nullable()->comment('% of eval weeks where direction class was correct');
|
||
$table->decimal('mae_pence', 5, 2)->nullable()->comment('Mean absolute error in pence × 100');
|
||
$table->json('calibration_table')->nullable()->comment('{bin_low..bin_high → empirical_hit_rate}');
|
||
$table->boolean('leak_suspected')->default(false)->comment('Secondary smell test: directional_accuracy > 75');
|
||
$table->dateTime('ran_at');
|
||
$table->timestamps();
|
||
|
||
$table->index(['ran_at']);
|
||
});
|
||
}
|
||
|
||
/**
|
||
* Reverse the migrations.
|
||
*/
|
||
public function down(): void
|
||
{
|
||
Schema::dropIfExists('backtests');
|
||
}
|
||
};
|