feat(forecasting): build calibrated weekly forecast stack with LLM overlay and volatility detector
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>
This commit is contained in:
@@ -21,17 +21,50 @@ final class FredBrentPriceSource
|
||||
*/
|
||||
public function fetch(): ?array
|
||||
{
|
||||
return $this->call([
|
||||
'sort_order' => 'desc',
|
||||
'limit' => 30,
|
||||
]);
|
||||
}
|
||||
|
||||
/**
|
||||
* Backfill range (inclusive). FRED's `observation_start` /
|
||||
* `observation_end` parameters expect ISO dates (YYYY-MM-DD).
|
||||
* Returns null when the range is empty (e.g. all weekends/holidays).
|
||||
*
|
||||
* @return array{date: string, price_usd: float}[]|null
|
||||
*
|
||||
* @throws BrentPriceFetchException
|
||||
*/
|
||||
public function fetchRange(string $from, string $to): ?array
|
||||
{
|
||||
return $this->call([
|
||||
'observation_start' => $from,
|
||||
'observation_end' => $to,
|
||||
'sort_order' => 'asc',
|
||||
'limit' => 100000,
|
||||
]);
|
||||
}
|
||||
|
||||
/**
|
||||
* @param array<string, scalar> $extraParams
|
||||
* @return array{date: string, price_usd: float}[]|null
|
||||
*
|
||||
* @throws BrentPriceFetchException
|
||||
*/
|
||||
private function call(array $extraParams): ?array
|
||||
{
|
||||
$params = array_merge([
|
||||
'series_id' => 'DCOILBRENTEU',
|
||||
'api_key' => config('services.fred.api_key'),
|
||||
'file_type' => 'json',
|
||||
], $extraParams);
|
||||
|
||||
try {
|
||||
$response = $this->apiLogger->send('fred', 'GET', self::URL, fn () => Http::timeout(30)
|
||||
$response = $this->apiLogger->send('fred', 'GET', self::URL, fn () => Http::timeout(60)
|
||||
->retry(3, 200, fn (Throwable $e) => $this->shouldRetry($e))
|
||||
->throw()
|
||||
->get(self::URL, [
|
||||
'series_id' => 'DCOILBRENTEU',
|
||||
'api_key' => config('services.fred.api_key'),
|
||||
'sort_order' => 'desc',
|
||||
'limit' => 30,
|
||||
'file_type' => 'json',
|
||||
]));
|
||||
->get(self::URL, $params));
|
||||
} catch (ConnectionException $e) {
|
||||
throw new BrentPriceFetchException("FRED connection failed: {$e->getMessage()}", previous: $e);
|
||||
} catch (RequestException $e) {
|
||||
|
||||
Reference in New Issue
Block a user