Modern Portfolio Theory Flashcards
Sharpe Ratio Limitations
- assumes returns follow a normal distribution, not true in reality (lots of price shocks)
- all volatility is penalised, even upside
- can be manipulated to boost the apparent risk-adjusted returns
Markovitz’s Diversification
- Combining securities with less-than-perfect positive correlation, p<1
- Reduce the risk of the portfolio without sacrificing the portfolio’s expected return
- Relation between correlation and risk level
Expected Return of a Portfolio
E(rp) = w1E(r1) + w2E(r2)
Variance of a portfolio’s return
σ2p = w12σ12 + w22σ22 + w32σ32 + 2w1w2σ12 + 2w1w3σ13 + 2w2w3σ23
Assumptions of MPT
- All investors have the same one-period time horizon
- All investors are risk-averse, and their goal is to maximise their expected utility
- All investors can borrow or lend at the risk free rate
- All investors choose portfolio following mean-variance model
- no tx costs
- no income taxes
Expected utility
Function of expected returns and variance
* E[u(R)] = f(E(R), σ2(R))
Global Minimum Variance Portfolio
Furthest left point on efficient frontier
Sharpe Ratio
E(rp) - rf / σp
CAPM
E(R) = rf + 𝛽i(E(Rm - rf)
CAPM Assumptions
- All investors have the same one-period time horizon
- All investors are risk-averse, and their goal is to maximise their expected utility
- All investors can borrow or lend at the risk free rate
- All investors choose portfolio following mean-variance model
- no tx costs
- no income taxes
- All investors have identical expectations about E(R), std dev and p for all securities
- All investors are price-takers
Beta
𝛽i = Covi,M / σ2M
Portfolio Beta
𝛽p = w1𝛽1 + w2𝛽2 + … + wn𝛽n
Security Market Line
graph of the CAPM
* Represents the relationship between an asset’s risk, measured by beta, and its expected return
Jensen’s Alpha
measures the risk-adjusted excess return earned by a security or portfolio over a given period
* α = (Ri - rf) - 𝛽i(RM - rf)
Fama-French Three-Factor Model
small stocks (and value stocks) outperform big stocks (and growth stocks)
* ri = rf + (E(rm) - rf)𝛽m + (SMB)𝛽smb + (HML)𝛽hml + ei
Criticisms of Fama-French Three-Factor Model
- some argue that the size and value factors are a result of data mining
- ignores momentum factor
Carhart Four-Factor Model
Considered momentum factor
* ri - rf = αi + (E(rm) - rf)𝛽m + (SMB)𝛽smb + (HML)𝛽hml + (UMD)𝛽umd + ei
Fama-French Five-Factor Model
Added profitability and investment factors on the previous model (Robust minus Weak, Conservative minus Aggressive)
* ri - rf = αi + (E(rm) - rf)𝛽m + (SMB)𝛽smb + (HML)𝛽hml + (RMW)𝛽rmw + (CMA)𝛽cma + ei
Treynor Ratio
reward-to-volatility ratio
* TRp = E(rp) - rf / 𝛽p
market portfolio
fully diversified portfolio of all risky assets
Capital Market Line
Special case of CAL.
* represents all possible combinations of risk-free assets and the market portfolio
Prospect Theory
individuals are more sensitive to potential losses than to equivalent potential gains
SMB
Small minus Big
* represents the difference in returns due to the small firm effect, small stocks have greater returns that big stocks
HML
High minus Low
* High B/M vs Low B/M, represents the difference in returns between value (High B/E) and growth (Low B/E) stocks