Asset Allocation Policy Recommendation
AFIN 839
Portfolio management
Major Assignment -Stage 3:
Asset Allocation Policy Recommendation
Semester 1 2015
Student number: 42408458
Student name: Wenhao Shi
Introduction
The main purpose of this assignment is that to analyze and compare the expected return and risk of portfolio with different portfolio weight used by four asset allocation policies, there are equally weight approach, Risk-minimizing approach, Mean-variance optimization approach and ‘Reward to risk timing’ approach. Because the underlying data is huge, therefore, I setting to analyze the data for 5 periods from 2001to 2014, then compare the relevant financial data, in order to find the best market portfolio as the benchmark in order to provide a recommended approach eventually.
The evaluation about four methods
An equally weighted investment approach
For this approach, we assumed that there are maintain a fixed percentage of six industries assets in the portfolio, so we set the each weight is 1/6, so these are not allow short selling in this approach, and we assumed the portfolio will be rebalanced annually categorized. Then we assumed that we calculate the sample mean, and then put the formula to calculate the expected returns of the portfolio for each 10 years date. The result is displayed below:
TABLE 1 | 2001-2010 | 2002-2011 | 2003-2012 | 2004-2013 | 2005-2014 |
Portfolio Mean | 0.527958333 | 0.537833333 | 0.785013889 | 0.832055556 | 0.836597222 |
Portfolio SD | 4.961374714 | 5.030285288 | 4.86171174 | 4.815309386 | 4.852858743 |
Portfolio SR | 0.070183841 | 0.077348164 | 0.133546356 | 0.146367658 | 0.148197298 |
A risk-minimizing approach
For the risk-minimizing approach, the first assumption for this approach is that the portfolio should be rebalance annually during the each period, then we use the single factor model to compute the covariance and variance.
In addition, because we need calculate the minimizing-risk portfolio, and the difference of risk of each industry, so I think the portfolio short selling must be allowed. However, when the portfolio short selling is not allowed, we still can get the result of the portfolio, then compare two results to find the better situation. So we also need assumed that there are no short selling constraints in this approach.
When we do the excel for this approach, we need find the covariance between each two assets, then calculate the variance and covariance matrix by this data with six industries asset, and then use the solver to calculate the best portfolio with the minimum variance. In addition, there are two situations need to be considered (short sales not allowed and short sales allowed).
There are showed two table in the below, Table2 (short sales not allowed) and Table 3(short sales allowed) show the results of the portfolio in two situations.
We comparing these tables, the standard deviation of portfolio in table 3 is lower than table 2, and sharpe-ratio of portfolio in table 3 is higher than table 2. There fore, when the short selling is allowed, the risk of portfolio is lower, and the portfolio is more valuable.
TABLE 2 | 2001-2010 | 2002-2011 | 2003-2012 | 2004-2013 | 2005-2014 |
Portfolio Mean | 0.352057488 | 0.569733235 | 0.825077563 | 0.872409968 | 0.960413543 |
Portfolio SD | 3.538969239 | 3.450078209 | 3.163286587 | 3.159628509 | 3.199528639 |
Portfolio SR | 0.048688609 | 0.12202136 | 0.217914989 | 0.235837842 | 0.263475334 |