Random Variable

What Is a Random Variable?

A random variable is a rule that assigns a number to each possible result of an uncertain event. The randomness lies in which result occurs, while the variable records the numerical value attached to that result. A coin flip is a simple example: if heads is worth 1 and tails is worth 0, the outcome is a random variable.

In finance, many quantities investors care about are random variables because their future values are not known in advance. Examples include tomorrow's closing price of a stock, the return on an index over the next year, the number of shares traded in a session, or the value of a portfolio on a future date.

Random variables are often grouped by the values they can take. A discrete random variable takes countable values, such as the number of up days in a month. A continuous random variable can take any value within a range, such as a daily percentage return.

How Random Variables Are Described

Every random variable is described by a probability distribution, which shows how likely each value or range of values is. Two summary measures are especially common:

• The expected value, or mean, is the probability-weighted average of all possible outcomes.
• The variance and its square root, the standard deviation, measure how widely outcomes spread around the mean.

For a discrete random variable, the expected value is calculated as:

Expected value = sum of (each possible value × the probability of that value)

Why It Matters and Its Limits

Describing uncertain financial outcomes as random variables lets analysts estimate typical results and risk. If a stock's annual returns are modeled as a random variable with a given mean and standard deviation, the model can estimate how often a loss of a certain size might occur. Many such models assume returns follow a normal distribution, a bell-shaped curve. Real market returns often show larger extreme moves than this assumption predicts, so results depend heavily on whether the chosen distribution fits the data.

The estimates are only as good as the historical data and assumptions behind them. A model fitted to past returns may not describe future conditions. Random variables describe uncertainty; they do not remove it.