Novus Methodology: How We Analyze Fund Performance
Read our latest article to learn how Novus analyzes fund performance, helping clients and delivering groundbreaking research.
Novus has consistently advised to clients to not rely on performance metrics alone for their manager evaluation process. In fact, we don’t recommend making an investment decision based on any one data point. Manager evaluation is a complex, nuanced endeavor as many of the industry’s participants can attest to. This is why we’re quickly becoming one of the largest repositories of manager position and exposure data and are actively looking for more ways to isolate manager skills and risks. However, performance is what you eat at the end of the day and it’s very important to have a simple tool that helps analyze returns for any one of your funds or portfolios. In this post we’ll explore the Novus performance tool set and some of the less-understood analytics used in assessing performance of managers. For data, we’ll use a hypothetical LSEQ fund we created just for demonstration purposes.
Basic Performance Analysis
To analyze the performance of a fund or portfolio, Novus clients simply click on the Performance tab to begin (1).
The rest of the settings are optional as they all have certain defaults, but you can select an appropriate benchmark (2), set the desired risk free rate (3), select monthly or daily data frequency, (4), and set the date range you wish to analyze (5). These settings save for all the work you’ll do on the performance tab. We realize the importance of selecting the right benchmark for the fund you’re analyzing. Novus has hundreds of indexes loaded including all the HFR hedge fund indexes.
The first thing we will do is switch to daily mode – (daily returns can be calculated from positions and P&L) daily returns are preferable to monthly numbers because they’ll provide truer measures of sensitivity and correlation.
The first sub-tab in the performance section is called summary and is a good starting point for your analysis. It looks like this:
The top half of the page is dedicated to statistical metrics that most people will find very familiar. The great thing about this though is that clicking on any of the metrics in blue will take you to a historical view of that metric. So for instance knowing that the Volatility of the fund is 10% over the whole period is great, but how has that changed recently? Clicking on the metric will take you to a rolling time window analysis for the selected metric and time period.
The second half of the page is shared by two charts, a simple scatter plot of risk vs. return and a bar chart of recent period returns compared to your benchmarks.
Next, clicking on the returns tab will bring up a cumulative returns chart with daily compounded returns:
A great feature of this cumulative performance chart is the ability to instantly change the vantage point by dragging on red line to any date you wish. Doing this will rest the returns to 0 for that point in time so you can quickly see when the investment has added value, and when it has stopped outperforming the benchmarks.
Time Window and Cone Plot Analysis
Rolling time window analysis is useful for investors who want to understand how a certain risk metric has changed for a manager over time. Take Standard Deviation, for instance – you know what the overall volatility for the fund has been over the recent years, but what is it now compared to a few years ago? By clicking on Time Window and selecting the desired period (180 days here) and the metric you want to analyze (Standard Deviation) you can really study of the volatility has evolved.
Another great feature of this analysis is the little-understood but very useful cone plot. Here is a different metric, the Sharpe Ratio (you can run this on any other metric as well).
The cone plot shows you where the current period’s Sharpe Ratio (or any metric) is on a relative basis for all such periods. The best way to explain the value of a cone plot is to absorb it in parts. First, we will double click on the “Latest” legend to only show one line – basically isolating all the day periods from today back.
Hovering over 252 for instance, we see that the Sharpe ratio for the most recent 252 days was 0.77 and moving the cursor along this line you can get a feel of all the other daily periods. How does this most recent 252-day window compare to the mean, max and minimum numbers for ALL 252-day windows in the managers history? We can answer that by turning on the Mean, Max and Min legends.
By hovering over the 252 day period now, you can see how it compares to the average of all such periods. In this case, it’s less than the average 252-day period has yielded. You can now turn on the lower and upper quartiles for even more context.
Next, we might choose to analyze drawdowns of the fund to understand how the manager was able to protect against downside volatility. Clicking on Draw Down will take you to that part of the performance analysis.
We can see here that the drawdowns on the long side of the book were fairly in line with those of the benchmark, save Q2 of 2014 where the fund was down significantly more than the benchmark. You can also use draw down analysis to understand how long it took a fund to get back to break even compared to the benchmark:
The Benchmark tab is where you would go for all your sensitivity metrics like Beta, Correlations as well as Alpha and Capture ratios. Keep in mind that everything in blue font allows you to click-through and view that metric historically in the time-window analysis.
Understanding and analyzing performance is important for institutional investors and managers alike. A good tool will allow on-the-fly analytics and great vitalization to answer the many questions that a sophisticate investor might have on their fund. Contact Novus to learn more about our performance analytics tools.
Three Managers. Three Styles.
An analysis of three different hedge fund strategies and how they generate alpha.
What makes a successful hedge fund manager? We believe that studying performance alone does not adequately answer the question. By looking at thousands of active manager portfolios, we peek under the hood of managers’ investment process and recognize patterns that lead managers to consistent outperformance. These patterns can be viewed as investment skill, or managers’ ability to generate alpha through certain repeatable methods.
In this article we use public ownership data and the Novus Alpha platform to evaluate three very distinct money managers. The three case studies aim to identify the driver of these managers’ success: investment skill.