# Macro & Quantitative Frameworks

## Dalio's Economic Machine

### Three Forces
1. **Productivity Growth** (~2% long-term trend) — Technology, education, infrastructure. Slow but stable.
2. **Short-Term Debt Cycle** (5-8 years) — Business cycle: expansion → peak → contraction → trough. Driven by credit availability.
3. **Long-Term Debt Cycle** (75-100 years) — Secular debt accumulation → deleveraging. Current position: identify which phase.

### Four-Box Framework (Asset Allocation by Regime)

|  | Rising Growth | Falling Growth |
|--|---------------|----------------|
| **Rising Inflation** | Stocks, Commodities, EM equities | Commodities, Gold, TIPS |
| **Falling Inflation** | Stocks, Nominal Bonds | Long-term Bonds, Cash |

Determine which box the current macro environment occupies. This drives the asset allocation overlay.

### Cycle Position Assessment
- Economy: Vigor or Slowing? (PMI > 50 = expansion; < 50 = contraction)
- Lenders: Eager or Restrained? (Credit spreads, lending standards survey)
- Spreads: Narrow or Wide? (HY OAS vs. historical range)
- Investors: Optimistic or Pessimistic? (AAII sentiment, put/call ratio)

If most answers left-column → defensive posture. Right-column → aggressive.

## Soros's Reflexivity Model

### 8 Phases
1. **Unrecognized Trend** — A fundamental trend exists but is not yet priced in
2. **Self-Reinforcing Phase** — Price movement confirms the trend, attracting more participants
3. **Successful Testing** — Corrections that would normally break the trend fail to do so
4. **Growing Conviction** — Participants become more confident; position sizes increase
5. **Flaw in Perceptions** — A gap between perception and reality widens but is ignored
6. **Climax** — The trend accelerates; latecomers pile in; divergence between price and fundamentals
7. **Reversal** — A catalyst exposes the flaw; trend reverses sharply
8. **Crash** — Self-reinforcing on the downside; forced selling

**Application**: For any stock, identify which reflexivity phase it occupies. Phase 1-3 → bullish. Phase 4-5 → cautious. Phase 6 → reduce. Phase 7-8 → avoid/short.

### Self-Reinforcing Loop Check
Is there a mechanism where:
- Rising stock price → improves fundamentals? (e.g., lower cost of capital, acquisition currency, talent retention via equity comp)
- Falling stock price → deteriorates fundamentals? (e.g., higher cost of capital, talent exodus, credit rating downgrade)

If yes, reflexivity is active — amplify conviction in the identified phase direction.

## Druckenmiller's Integration Framework

- **Macro is the driver, micro is the vehicle**: Get the macro right first. The best company in a bad sector still struggles.
- **Liquidity focus**: Central bank direction is the single most important variable. Is the Fed/ECB/PBoC easing, neutral, or tightening?
- **Position sizing by conviction**: 10-50% of portfolio for high-conviction ideas. Size inversely to uncertainty.
- **Kill switch**: Define what makes you wrong BEFORE entering. Write it down. When triggered, exit without hesitation.

### Druckenmiller Entry Checklist
1. Is liquidity favorable (central bank easing or neutral)?
2. Is the macro backdrop supportive for this sector?
3. Is there a catalyst within the next 3-6 months?
4. Is the risk/reward ratio >3:1?
5. What specifically would prove me wrong?

## The Kelly Criterion (Mathematical Sizing)

Used to determine the optimal position size based on the probability of success and the risk/reward ratio. To account for estimation error, the agent should apply **Fractional Kelly (1/4 or 1/2)**.

### The Formula
`f* = (p * b - q) / b`

- `f*`: Fraction of portfolio to allocate
- `p`: Probability of the Bull/Base scenario (from Stage 7.3)
- `b`: Net odds (Upside to Target / Downside to Stop Loss)
- `q`: Probability of the Bear scenario (1 - p)

### Application Rules
1. **Never use Full Kelly**: It leads to extreme volatility. Default to **Quarter Kelly (0.25 × f*)**.
2. **Hard Cap**: Regardless of Kelly output, never exceed Druckenmiller's **50% cap** for a single idea.
3. **Negative Kelly**: If the formula returns a negative number, the "edge" is absent — the recommendation must be **Avoid/Sell**.

## Greenblatt's Magic Formula

Rank stocks by two factors combined:
1. **Earnings Yield** = EBIT / Enterprise Value (higher = cheaper)
2. **Return on Capital** = EBIT / (Net Working Capital + Net Fixed Assets) (higher = better business)

Buy top 20-30 combined-rank stocks, hold 1 year, rebalance.

### Application to single-stock analysis
Compute the company's earnings yield and ROC. Compare to:
- Magic Formula universe median (if available)
- Sector median
- Company's own 5-year average
- S&P 500 median (~5% earnings yield, ~20% ROC approximate)

A company in the top quartile on both metrics is attractive on quantitative grounds alone.