Five stages of the QIWU process
- Observe — Identify the available market, probability or behavioural inputs without forcing a conclusion.
- Structure — Organise those inputs into a consistent framework.
- Simulate — Explore more than one possible path using stated assumptions.
- Compare — Review risk, invalidation, alternatives and limitations.
- Decide and review — The user makes the final decision and later reviews the process, not only the result.
Market-analysis framework
AI 阿公 GPT uses seven recurring analytical pillars. The framework produces observations and scenarios. It does not turn a chart into certainty.
- Market Structure — HH, HL, LH, LL, BOS and CHoCH.
- Trend & Consolidation — Directional conditions, range behaviour and changing momentum.
- Support & Resistance — Areas where price previously reacted or may attract attention.
- Price Action — Candle behaviour around important areas.
- Volume–Price Relationship — Whether activity supports or conflicts with the price move.
- Money Flow Analysis — Available evidence of participation and directional pressure.
- Liquidity Sweep — Possible stop hunts, liquidity grabs, inducement and failed breaks.
A mathematically positive model can still experience losses, drawdowns and long losing sequences. Expected value is not a promise of profit.
Probability framework
QIWU Probability Lab separates variables that are often confused: win rate, risk–reward ratio, expected value, sample size, sequence risk, drawdown and position sizing.
The simulation illustrates possible consequences under selected assumptions. It does not diagnose a person or predict an actual financial future.
Behavioural framework
2036 Future Experience examines the repeated actions behind the numbers.
- Trading or investing frequency
- Risk per decision and stop-loss discipline
- Averaging down, adding to a position and leverage
- Responses to gains, losses and drawdowns
- FOMO, loss aversion and short-term reward bias
The role of AI
AI can organise information, annotate charts, apply a defined framework and generate alternative scenarios quickly. It can also misunderstand an image, miss context, apply a rule inconsistently or produce incorrect information.
For this reason, QIWU treats AI as a decision-support layer—not as an autonomous authority.
Methodology governance
- State the methodology version and review date on evidence pages.
- Separate historical, simulated and hypothetical material clearly.
- Preserve original inputs wherever a case review is presented.
- Document invalidations, errors and limitations—not only favourable outcomes.
- Update public claims when the underlying method or sample changes.