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>
57 lines
1.8 KiB
PHP
57 lines
1.8 KiB
PHP
<?php
|
|
|
|
use App\Services\Forecasting\BacktestRunner;
|
|
use App\Services\Forecasting\Models\NaiveZeroChangeModel;
|
|
use Carbon\Carbon;
|
|
use Illuminate\Foundation\Testing\RefreshDatabase;
|
|
use Illuminate\Support\Facades\DB;
|
|
|
|
uses(RefreshDatabase::class);
|
|
|
|
it('predicts zero change with flat direction', function () {
|
|
$model = new NaiveZeroChangeModel;
|
|
|
|
$prediction = $model->predict(Carbon::parse('2024-06-03'));
|
|
|
|
expect($prediction->magnitudePence)->toBe(0.0)
|
|
->and($prediction->direction)->toBe('flat');
|
|
});
|
|
|
|
it('has an empty FeatureSpec (no features by design)', function () {
|
|
$model = new NaiveZeroChangeModel;
|
|
|
|
$spec = $model->featureSpec();
|
|
|
|
expect($spec->modelLabel)->toBe('naive-zero')
|
|
->and($spec->features)->toBe([])
|
|
->and($spec->modelVersion())->toStartWith('naive-zero-');
|
|
});
|
|
|
|
it('runs cleanly through the backtest harness on real-shape data', function () {
|
|
// 8 weeks gently rising — naive predicts flat → expect 0% accuracy.
|
|
$start = Carbon::parse('2024-01-01');
|
|
for ($i = 0; $i < 8; $i++) {
|
|
DB::table('weekly_pump_prices')->insert([
|
|
'date' => $start->copy()->addWeeks($i)->toDateString(),
|
|
'ulsp_pence' => 14000 + ($i * 100),
|
|
'ulsd_pence' => 15000 + ($i * 80),
|
|
'ulsp_duty_pence' => 5295,
|
|
'ulsd_duty_pence' => 5295,
|
|
'ulsp_vat_pct' => 20,
|
|
'ulsd_vat_pct' => 20,
|
|
]);
|
|
}
|
|
|
|
$result = (new BacktestRunner)->run(
|
|
new NaiveZeroChangeModel,
|
|
trainStart: Carbon::parse('2024-01-01'),
|
|
trainEnd: Carbon::parse('2024-01-29'),
|
|
evalStart: Carbon::parse('2024-02-05'),
|
|
evalEnd: Carbon::parse('2024-02-19'),
|
|
);
|
|
|
|
expect((float) $result->directional_accuracy)->toBe(0.0)
|
|
->and((float) $result->mae_pence)->toBe(1.0)
|
|
->and($result->leak_suspected)->toBeFalse();
|
|
});
|