Chap027博迪凱恩馬庫斯《投資學》課件

1、CHAPTER 27The Theory of Active Portfolio ManagementOverviewTreynor-Black modelOptimization using analysts forecasts of superior performanceAdjusting model for tracking errorAdjusting model for analyst forecast errorBlack-Litterman modelTable 27.1 Construction and Properties of the Optimal Risky Port

2、folioTable 27.2 Stock Prices and Analysts Target Prices for June 1, 2006Figure 27.1 Rates of Return on the S&P 500 (GSPC) and the Six Stocks, June 2005 May 2006 Table 27.3 The Optimal Risky Portfolio with the Analysts New Forecasts Table 27.4 The Optimal Risky Portfolio with Constraint on the Active

3、 Portfolio (WA 1) Figure 27.2 Reduced Efficiency when Benchmark is Lowered Table 27.5 The Optimal Risky Portfolio with the Analysts New Forecasts (benchmark risk constrained to 3.85%)Adjusting Forecasts for the Precision of AlphaHow accurate is your forecastHow should you adjust your position to tak

4、e account of forecast imprecisionMust quantify the uncertainty by examining the forecasting record of previous forecasts by same forecasterThe adjusted alpha:Figure 27.3 Histogram of the Alpha ForecastFigure 27.4 Organizational Chart for Portfolio Management Steps in the Black-Litterman ModelStep 1:

5、 Estimate the covariance matrix from historical dataStep 2: Determine a baseline forecastStep 3: Integrating the managers private viewsStep 4: Developing revised (posterior) expectationsStep 5: Apply portfolio optimization Figure 27.5 Sensitivity of Black-Litterman Portfolio Performance to Confidenc

6、e Level (view is correct) Figure 27.6 Sensitivity of Black-Litterman Portfolio Performance to Confidence Level (view is false)The BL Model as Icing on the TB CakeSuppose that you have two portfoliosone for the US and one for EuropeThe model would be run as two separate divisionsEach division would c

7、ompile values of alpha relative to their own passive portfolioRelative performance of the two markets can be expected to add information to the independent macro forecasts for the two economiesPortfolios need to be optimized separatelyValue of Active ManagementModel for estimation of potential feesK

8、ane, Marcus, and Trippi derive an annuitized value of portfolio performance measured as a percent of funds under managementThe percentage fee that investors would be willing to pay for active services can be related to the difference between the square of the portfolio Sharpe ratio and that of the p

9、assive portfolioSource of the power of the active portfolio is the additive value of the squared information ratiosTable 27.6 M-Square for the Portfolio, Actual ForecastsTable 27.7 M-Square of Simulated PortfoliosConcluding RemarksThe gap between theory and practice has been narrowing in recent yearsThe CFA is expanding knowledge bas