Portfolio Management 2/19

Introduction to Portfolio Theory

  • Harry Markowitz is the central figure in the development of portfolio theory, having introduced key concepts in the 1950s.

  • Despite not being conventionally attractive, Markowitz won the Nobel Prize for his contributions to finance, specifically for the concept of efficient portfolios.

Efficient Portfolios

  • Definition: Efficient portfolios are designed to provide the maximum possible return for a given level of risk or the minimum risk for a given level of return.

  • No portfolio can outperform the efficient frontier — a graphical representation of the best possible portfolio combinations.

  • Graphical Representation: In discussing the efficient frontier, Markowitz plotted risk on the y-axis and return on the x-axis, which was counterintuitive, as one would typically expect the risk-return relationship to be inverse.

Key Concepts in Portfolio Theory

  • The return of a portfolio: Linear combination of individual securities.

    • Formula: If a portfolio consists of multiple securities, the return can be expressed as:
      E(R<em>portfolio)=w</em>1E(R<em>1)+w</em>2E(R<em>2)+…+w</em>nE(R<em>n)E(R<em>{portfolio}) = w</em>1 E(R<em>1) + w</em>2 E(R<em>2) + … + w</em>n E(R<em>n) where (wi) denotes the weight of each security in the portfolio and (E(R_i)) denotes the expected return of each security.

  • The variance of a portfolio: Not a linear combination of individual security variances.

    • Key assertion: Variance depends on both weights of the securities and their covariances.

    • Formula for the variance of a two-security portfolio:
      Var(R<em>portfolio)=w</em>12Var(R<em>1)+w</em>22Var(R<em>2)+2w</em>1w<em>2Cov(R</em>1,R<em>2)Var(R<em>{portfolio}) = w</em>1^2 Var(R<em>1) + w</em>2^2 Var(R<em>2) + 2 w</em>1 w<em>2 Cov(R</em>1, R<em>2) where (Cov(R1, R_2)) represents the covariance between the returns of securities 1 and 2.

Mathematical Foundations of Portfolio Variance

  • For a portfolio with multiple securities, the variance includes correlations and covariances:

    • The expression grows complex as the number of securities (n) increases, requiring consideration of all pairs of securities.

  • Important Variables:

    • (
      ho ) (correlation coefficient) impacts the portfolio risk.

  • As the correlation between securities decreases, the overall portfolio risk decreases, therefore emphasizing the idea that diversification reduces risk.

Minimum Variance Portfolio

  • Definition: The portfolio that has the least risk (variance) for a given level of expected return.

  • Allocation in Two-Security Portfolio: To find the minimum variance, we set weights and derive using derivatives in calculus.

    • Optimal allocation: If (Xa) is the weight in security A, we find: X</em>b=1−X<em>aX</em>b = 1 - X<em>a which allows us to express the portfolio variance in terms of (Xa).

  • Deriving minimum variance involves taking derivatives of variance function,
    followed by setting the derivative to zero, and leading to optimal weight allocations:
    X<em>a=racVar(R</em>b)−Cov(R<em>a,R</em>b)Var(R<em>a)+Var(R</em>b)−2Cov(R<em>a,R</em>b)X<em>a = rac{Var(R</em>b) - Cov(R<em>a, R</em>b)}{Var(R<em>a) + Var(R</em>b) - 2Cov(R<em>a, R</em>b)}
    where Xb follows as (1 - Xa).

Risk and Return Trade-off

  • The central premise of portfolio management is to reduce risk for a given level of return, rather than solely maximizing returns.

  • Total Risk is considered, and investors seek the minimum variance portfolio that yields the targeted return.

  • The importance of correlation in diversifying is underlined — perfectly negatively correlated securities will yield higher diversification benefits, whereas perfectly positive correlated securities will not reduce portfolio risk effectively.

Single Index Model

  • Introduced to simplify the calculations associated with portfolio variance, comparing each security to a benchmark (e.g., the market portfolio).

  • Reducing Complexity: Instead of calculating covariance between every pair of securities (which grows exponentially), the single index model allows for a simpler analysis by focusing on how each security reacts relative to the index.

  • The Beta of a security is a key measure derived from this model, reflecting its risk in relation to market variability.

    • The formula for beta:
      β=racCov(R<em>i,R</em>m)Var(R<em>m)\beta = rac{Cov(R<em>i, R</em>m)}{Var(R<em>m)} where (Ri) is the return of the individual stock and (R_m) is the return of the market portfolio.

Implications of the Single Index Model

  • The model drastically reduces the number of computations necessary to calculate portfolio risk and return.

  • The beta of a portfolio can be similarly determined using the weights of the individual securities' betas:
    β<em>portfolio=w</em>1β<em>1+w</em>2β<em>2+…+w</em>nβn\beta<em>{portfolio} = w</em>1\beta<em>1 + w</em>2\beta<em>2 + … + w</em>n\beta_n

  • This means that understanding the overall risk profile of a portfolio becomes more manageable without delving into extensive calculations of variances and covariances.

Conclusion

  • Harry Markowitz’s work laid the foundation for modern portfolio theory, emphasizing the importance of diversification, risk reduction, and the mathematical formulations that yield efficient portfolios.

  • Understanding and applying these concepts is crucial for successful investment management and risk assessment, aiming not merely for high returns but also for risk mitigation through strategic portfolio composition.