ai trading / options trading / walk forward testing / backtesting
AI Options Strategy: Build With Aurora, Test on Unseen Data
Give Aurora a watchlist screenshot and inspect the strategy she builds. This workflow shows the entry and exit rules, compares four position sizes using historical backtests, and evaluates selected settings across three later walk-forward test windows.
The final steps attach the selected rules to an existing Public brokerage account. Its imported account history is labeled separately from the new rule attachment. You can also paper-trade with simulated money before connecting a supported brokerage.
Transcript
0:00Turn your watchlist into a testable strategy.
0:03Build an options strategy for this watchlist. Describe Attachments: 24 symbols.
0:06Generate Portfolios. Backtest Portfolios. Compare Backtests. Four position sizes: 3%, 5%, 7%, 10% per stock.
0:10Inspect the selected rules: long-dated calls. Price above the 100-day average and positive 63-day return; close when both reverse.
0:14Walk-Forward Study. Three historical test windows. Anchored walk-forward with a 14-day embargo. Strategy compared with SPY total return. Each test starts from $25,000.
0:21Choose a brokerage. Public, Robinhood Agentic, and TradeStation.
0:26Attach strategies to an existing Public brokerage account. Confirm and port.
0:31Public Portfolio Challenge. Existing account history starts at $25,000. Imported brokerage history before this rule attachment. nexustrade.io.
Join the conversation
Loading conversation…