Monday, 29 August 2011

CLUSTER ANALYSIS



Applications in the field of Portfolio management

Cluster analysis can reveal important insights into portfolio management behaviour during the financial crisis of 2008–2009. Fund managers always used Quantitative methods include the calculation of the Sharpe ratio and other performance measures, portfolio optimization, and analysis of the correlation matrix of returns. Using cluster analysis can provide further insights into the pattern followed by these fund managers. Various clusters can be formed using variables such as asset class, style of hedge fund, returns, incentive fee, risk level, and liquidity. Here point to be noted is that unlike in case of the examples solved in class, we need to classify the funds and not the variables. Based on this objective we can gather data on 50 funds asking questions as follows:-

Asset class

1) % of investment in equity

2) % of investment in money market instruments

3) %of investments in bonds

Incentive fee

1) % of variable component in the incentives realised by the fund managers

2) % of fixed component in the incentives realised by the fund managers

Returns

1) % of return over last 3 months

2) % of returns over last one year

3) % of returns over last 3 years

4) % of returns over last 5 years

As all the variables are of the interval measurement, the funds can be clustered and analysed further using dendogram, agglomeration schedules and proximity matrix.

After the clusters have been formed we derive various conclusions

For example

- Direct Relationship between incentive fee and returns

- Exposure to asset class varying with the fund strategy of the fund manager

These are direct observations, but we need to go one step further and create a profile to interpret the cluster analysis done by SPSS. For this, certain other variables need to be looked upon to dig deeper such as macroeconomic factors, competitor’s risk-return profile, autonomy of the fund manager, etc.

Based on these observations and interpretations various recommendations can be given to create the efficient portfolio which maximises the investor’s returns.

Author: Asmita Karanje

Group: Finance 3

No comments:

Post a Comment