Overview run_trading_job is the final step of the core pipeline . It triggers the trading-worker pipeline job, which runs a VectorBT backtesting simulation using the optimized portfolio weights and returns performance metrics.The tool blocks until the job completes and returns the output blob URL.
Parameters Blob storage URL pointing to a portfolio_optimization_*.json file. This is the output_url returned by run_po_job.
Trading and backtesting configuration. Backtest execution settings. Show Trading params fields
Backtesting engine. Supported values:
"VectorBT" — weights-based backtest (default)"VectorBT_Advanced" — weights-based with extended options (VectorBT params block)"VectorBT_from_signals" — signal-based; requires Entries and Exits DataFrames in the input JSON Initial portfolio cash in USD. Default: 1000000.
Rebalancing frequency as a pandas Timedelta unit.
Examples: "D" (daily), "h" (hourly), "min" (minute). Default: "D".
Transaction cost as a fraction of trade value (not basis points).
Example: 0.0003 = 3 bps. Default: 0.0003.
Slippage in basis points. Example: 3 = 3 bps. Default: 3.
Annual risk-free rate used for Sharpe/Sortino ratio calculations. Default: 0.0.
Advanced VectorBT options. Applied only when Backtest method is "VectorBT_Advanced". Show VectorBT params fields
How portfolio sizes are interpreted. Options: "TargetPercent" (default),
"TargetValue", "TargetShares", "Amount".
Share cash pool across all assets. Default: true.
Group assets for portfolio-level statistics. Default: true.
Ticker symbol for benchmark comparison. Default: "SPY".
Date range for the backtest evaluation window. Show Backtest params fields
Start of the full dataset. Format: YYYY-MM-DD.
End of training / start of out-of-sample testing. Backtest begins one business day after this date.
Format: YYYY-MM-DD.
End of the out-of-sample testing period. Format: YYYY-MM-DD.
Returns { "status" : "Succeeded" , "output_url" : "https://stmcpfabricdev.blob.core.windows.net/data/trading_report_20260307_130504.json" , "output_name" : "trading_report_20260307_130504.json" , "execution_name" : "trading-worker-abc123xyz" }
Field Description statusJob terminal status (Succeeded) output_urlFull HTTPS URL to the backtest results blob output_nameBlob filename execution_nameJob execution ID for audit/debugging
The trading_report_*.json blob contains VectorBT performance metrics: total return, Sharpe ratio,
max drawdown, win rate, and per-ticker statistics.
Example — Standard VectorBT { "input_url" : "https://stmcpfabricdev.blob.core.windows.net/data/portfolio_optimization_20260307_130215.json" , "config" : { "Trading params" : { "Backtest method" : "VectorBT" , "initial_capital$" : 1000000 , "frequency" : "D" , "trade_comission" : 0.0003 , "slippage_bps" : 3 , "risk_free_rate" : 0.02 }, "Backtest params" : { "Learning_start" : "2020-01-01" , "Learning_end" : "2024-01-01" , "Testing_end" : "2026-03-01" } } }
Example — VectorBT Advanced { "input_url" : "..." , "config" : { "Trading params" : { "Backtest method" : "VectorBT_Advanced" , "initial_capital$" : 500000 , "frequency" : "D" , "trade_comission" : 0.001 , "slippage_bps" : 5 , "risk_free_rate" : 0.04 , "VectorBT params" : { "size_type" : "TargetPercent" , "cash_sharing" : true , "group_by" : true , "benchmark_ticker" : "SPY" } }, "Backtest params" : { "Learning_end" : "2024-01-01" , "Testing_end" : "2026-03-01" } } }
Resources Resource Value Container Apps Job trading-workerContainer name trading-workerEnv vars injected INPUT_URL, CONFIGOutput blob prefix trading_report_Timeout 600 seconds
Next Steps This is the last step of the core pipeline. Download the blob from output_url to inspect the results.
The trading_report_*.json file contains VectorBT performance metrics: total return,
Sharpe ratio, max drawdown, win rate, and per-ticker statistics.
You can also run optional post-processing tools from portfolio_optimization_*.json:
