What are the key ex-post measures necessary to perform effective risk analysis on a multi-asset portfolio?
1. Regression Analysis (Covariance, Correlation, R2, …)
— Correlation Squared- Measure of the amount of portfolio’s performance that the movement of its benchmark can explain.
2. Benchmark Statistics
— Active Share: Measure of how different a portfolio’s holdings are relative to the holdings of the particular benchmark considered (A value close to zero shows fund holdings to have the same securities as that in the benchmark and thus the portfolio is passive and same returns can be achieved by holding the benchmark)
— Up/Down Capture
3. Relative Risk Measures
— Tracking Error
— Information Ratio: Risk-adjusted returns relative to benchmark )
4. Risk Ratios
— Standard Deviation – Fundamental risk measure that determines total risk. This is valuable when looked at it along with portfolio returns.
— Sharpe Ratio: Return over the risk-free rate per unit of total risk (standard deviation)
— Treynor Ratio: Return over the risk-free return for a given level of market risk (beta).
— Jensen’s Alpha: Excess returns compared to returns suggested by the CAPM model
5. Other Analytics (Kurtosis, Skewness, Drawdown, Sortino Ratio, …)
— Skewness – Measure of the asymmetry of the distribution of return about their mean. For a normal distribution, this value is zero and we know asset returns have non-zero skew.
— Kurtosis – Measure of the “peakedness“ of the distribution of returns i.e. it describes the size of the tails of the return distribution. Usually, asset returns have fat-tails with kurtosis is > 3.
— Sortino Ratio – It extends Sharpe ratio in that it also determines the risk-adjusted return but only penalizes the returns, which have downside risks.
Bringing all this data into a BI data store (such as snowflake/Redshift) along with holdings and performance analytics will be necessary for real-time view to make the analysis useful (See below image for a sample ex-post data set).
As has been the key theme through the posts related to real-time BI Analysis – use of modern ‘cloud’ technology using a microservices architecture that can scale dynamically depending on data volume.
This will enable firms to operate using the ‘Pay Per Use’ business model – a key competitive differentiator in the long-term for deeper actionable insights.