allucanget/arbitrade
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: enhance backtesting panel with flash messages and pairing checks
CI / lint-test-build (push) Failing after 12s
Details

Browse Source
This commit is contained in:
zwitschi
2026-06-07 21:51:09 +02:00
parent 5e7732b85f
commit f221464daa
4 changed files with 269 additions and 79 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/architecture/README.md
+10 -7
View File
@@ -53,7 +53,7 @@ The bot consumes Kraken market data, detects opportunities, and executes trades
- `detection/` - triangular graph and incremental detector.
- `risk/` - pre-trade and trade-limit guards.
- `execution/` - multi-leg trade sequencing.
- `backtesting/` - replay engine, parameter sweep, experiment scaffolds.
- `backtesting/` - replay engine, parameter sweep, experiment scaffolds. See [backtesting.md](backtesting.md).
- `strategy/` - experimental strategy modules such as stat-arb.
- `storage/` - PostgreSQL schema and repositories.
- `alerting/` - multi-channel notifications.
@@ -77,7 +77,7 @@ The bot consumes Kraken market data, detects opportunities, and executes trades
3. Incremental detector scores impacted cycles.
4. Risk manager validates the opportunity.
5. Execution sequencer places legs if approved.
7. Trades and snapshots persist to PostgreSQL.
6. Trades and snapshots persist to PostgreSQL.
7. Dashboard and alerts reflect state changes.
### 6.2 Dashboard Control Flow
@@ -89,11 +89,14 @@ The bot consumes Kraken market data, detects opportunities, and executes trades
### 6.3 Backtesting Flow
1. User selects JSONL replay file and run parameters.
2. Replay engine loads ordered book events.
3. Detector, risk, and execution logic run in simulation mode.
4. Report is stored in memory for recent UI display.
5. Parameter sweeps split data into train/test windows, rank results, and flag overfit.
See [backtesting.md](backtesting.md) for full design and implementation details.
1. User picks currency pairs (from config/pairings page, or all enabled).
2. User sets starting balances (required), time range (required), min profit threshold (required).
3. Fee profile defaults to "api (from Kraken)"; slippage (4.0 bps) and execution latency (20 ms) are optional with sensible defaults.
4. Job is queued via `POST /dashboard/backtesting/run`.
5. Backend loads events from `market_snapshots` table, builds triangular cycles, runs replay engine.
6. Report stored in `backtest_jobs` table, visible in recent jobs list.
## 7. Deployment View