finfund › ความรู้ › Portfolio Theory & Risk Management

Portfolio Theory & Risk Management

📊 Modern Portfolio Theory (MPT)

Markowitz won the Nobel Prize for this concept in 1990. The core idea: "diversify to reduce risk without reducing returns." The key is to choose assets that are not correlated — when one falls, another may rise or stay flat.

📈 Efficient Frontier

A graph showing "most efficient" portfolios — the highest possible return for each level of risk. Portfolios below this line are "inefficient" (same risk, lower returns).

🔀 Diversification

"Don't put all your eggs in one basket" — the most basic rule, but many get it wrong:

  • Asset class diversification: stocks + bonds + gold + real estate
  • Geographic diversification: Thailand + US + emerging markets
  • Sector diversification: tech + banking + energy + tourism
  • Time diversification: DCA monthly, not a single lump sum

⚠️ Diworsification: over-diversifying = high fees but market-average returns. 10-20 assets is enough.

🎲 Risk Measurement

  • Standard Deviation (σ) — measures volatility; higher = riskier
  • Beta (β) — vs. the overall market; β=1 matches market, β>1 is more volatile
  • Sharpe Ratio — return per unit of risk = (Return − Risk-free rate) ÷ σ; higher is better
  • Maximum Drawdown — deepest drop from peak; measures real "pain"

⚖️ Asset Allocation

Asset allocation determines 90% of returns (Brinson study) — more important than stock picking. Popular rules of thumb:

  • Rule of 110 minus age: equity % = 110 − your age (e.g. age 30 = 80% stocks)
  • Rule of 100 minus age: classic version (age 30 = 70% stocks)
  • 60/40: 60% stocks, 40% bonds (classic)
  • All-Weather (Ray Dalio): 30% stocks, 55% bonds, 15% gold/commodities

🎯 Kelly Criterion

A formula for optimal position sizing when you have an edge:

Kelly % = W − (L ÷ W)

Where W = win rate, L = loss rate. Example: win 60%, gain 2x when winning, lose 1x when losing → Kelly = 0.6 − (0.4 ÷ 2) = 0.4 = invest 40% of portfolio. But real investors use Half-Kelly for safety.

📉 Value at Risk (VaR)

Measures the "maximum expected loss" at a 95% confidence level over a period. Example: "95% chance the portfolio won't lose more than 50,000 baht in a month."

🔄 Rebalancing

When asset allocation drifts from target (e.g. stocks grew to 80% from a 60% target) — sell the excess, buy what's below target. Do every 6-12 months, or when drift exceeds 5%.