Quant strategy backtests for Japanese stocks
If you ask an AI whether a trade will make money, the verdict for that question is here. Passing a backtest and printing money are different things.
Buy what already went up. Buy after earnings. Bet that a drop bounces. Let a model pick the names. Those are the questions people type into a chatbot. qbuntu publishes the median across the parameter grid for each one. Read it yourself, or let your AI read the same numbers.
Before you ask an AI
common prompts, answered with the median verdictA record here is live or promising. That verdict is separate from printing money in production. Read it with the methodology.
トヨタ自動車 (7203.T) has no buy or sell rating here. The row records what happened when a tested model selected the name. Median long hit rate is 0.42 (n=10), below a coin flip. Median average long return is -25.4 bps.
The published records are rejected. A strong cell at the edge of the grid is not the answer to this question.
The same rule gets a different verdict on a different universe.
Quant analysis by stock
model–stock fit, not a buy signalClosest to the bar
verdict first, then median Sharpe · promising needs return and significanceFor AI agents
qbuntu speaks your language. Each strategy has a Markdown mirror at /ai/strategies/<slug>.md and JSON at /v1/strategies/<slug>. Cite atomic facts via /cite/<fact-id>. Connect via MCP at mcp.qbuntu.ai.
curl https://qbuntu.ai/v1/strategies/amihud-illiquidity-premium-jp-all