To analyze a hedge fund's performance, you first need to make sense of its track record. Imagine you have a list of a fund's daily returns for the past year—hundreds of data points. How do you boil that down to something useful? This is where descriptive statistics come in. They are tools for summarizing data into a few meaningful numbers.

The most common starting point is the mean, or average return. It gives you a sense of the fund's typical performance on any given day. You calculate it by adding up all the daily returns and dividing by the number of days.

But the mean can be misleading if there are extreme outliers. A single day with a massive gain or loss can skew the average. That's why we also look at the median, which is the middle value when you line up all the returns in order. It's often a better indicator of the true central tendency.

Of course, knowing the average return is only half the story. Two funds can have the same average return but be wildly different investments. One might be a steady, consistent earner, while the other is a rollercoaster of highs and lows. To capture this, we use the standard deviation ($\sigma$). It measures how spread out the returns are from the average. A low standard deviation means returns are clustered tightly around the mean—it's a less volatile, more predictable fund. A high standard deviation signals greater risk and volatility.

Asset returns aren't just random numbers; they often follow a pattern. A probability distribution is a mathematical function that describes the likelihood of different outcomes. For a long time, financial analysts have used the normal distribution, also known as the bell curve, to model investment returns.

The normal distribution suggests that most returns will be close to the average, and extreme gains or losses are very rare. According to this model, about 68% of returns fall within one standard deviation of the mean, and 95% fall within two. While this is a useful starting point, real-world financial returns often have "fat tails." This means that extreme, unexpected events happen more frequently than a perfect normal distribution would predict. Understanding this limitation is key to proper risk management.

Suppose a hedge fund claims its new trading algorithm generates an average monthly return of 1%. How can you be sure this isn't just due to luck? This is a job for hypothesis testing. It provides a formal framework for making decisions based on data.

You start with two competing statements:

1. The Null Hypothesis ($H_0$): This is the default assumption, the status quo. In our case, it would be that the algorithm has no effect, and the average monthly return is zero.
2. The Alternative Hypothesis ($H_a$): This is what you're trying to prove. Here, it would be that the algorithm does have an effect, and the average monthly return is not zero (or is greater than zero).

We analyze our data to calculate a p-value. The p-value is the probability of observing our data (or something more extreme) if the null hypothesis were true. A small p-value (typically less than 0.05) suggests that our observed results are very unlikely to have happened by random chance. This gives us the confidence to reject the null hypothesis and conclude that the fund's algorithm likely does have a real effect.

Hedge fund returns don't exist in a vacuum. They are influenced by many factors, such as the overall stock market, interest rates, or commodity prices. Regression analysis is a statistical tool that helps us understand and quantify the relationship between variables.

The simplest form is a linear regression, which tries to fit a straight line through a set of data points. For example, we could plot a hedge fund's monthly returns against the monthly returns of the S&P 500 index.

In this model, two coefficients are crucial:

• Beta ($\beta$): This measures the fund's sensitivity to the market. A beta of 1 means the fund tends to move in line with the market. A beta greater than 1 means it's more volatile than the market, and a beta less than 1 means it's less volatile.
• Alpha ($\alpha$): This is the intercept of the regression line. It represents the portion of a fund's return that is not explained by the market's movement. A positive, statistically significant alpha is often considered the holy grail for a portfolio manager, as it can indicate skill in generating returns independent of the broader market.

Let's check your understanding of these core statistical concepts.

These statistical tools—descriptive statistics, distributions, hypothesis testing, and regression—are the building blocks for any serious analysis of investment performance. They allow you to move beyond gut feelings and make data-driven judgments about risk and return.