91Ó°ÊÓ

When working with probability distributions, the expected value is a fundamental concept that helps us understand long-term averages. In simple terms, it represents the mean of a random variable across many trials.
To find the expected value, also denoted as \( \mu \), we multiply each possible outcome of a random variable by its probability, then sum all these values. For example, if we have a random variable \( y \) with a certain probability distribution as shown above, the expected value \( \mu_y \) is calculated as follows:

• Multiply each outcome \( y \) by its probability \( p(y) \).
• Add these products together.

This computation results in the average number of broken eggs you would expect per carton in the long run. Unlike a simple arithmetic average, it accounts for the probabilities of each result, providing a weighted average that reflects real-world scenarios more accurately. That’s why, in our example, even though the simple average of broken eggs is 2.0, the expected value is actually 0.56.

In probability theory, a random variable is a variable that can take on different values based on the outcome of a random event. Its value is not certain but can be described in terms of probabilities.
Random variables can be discrete, like in our egg carton example, where values are distinct and countable such as 0, 1, 2, 3, or 4 broken eggs. Each value corresponds to a probability that tells us how likely each outcome is to occur.

• The random variable \( y \) is defined as the number of broken eggs.
• Its possible values are integers corresponding to the number of broken eggs in a carton.
• The probability \( p(y) \) expresses how frequently each scenario happens.

Understanding random variables is crucial for calculating expected values and interpreting the results in a meaningful way. It provides the foundational basis for constructing probability distributions.

Probability weights play a critical role in determining the expected value of a random variable. These weights are essentially the probabilities assigned to each possible outcome and indicate the relative frequency or likelihood of each occurring.
In the context of our example, the following steps explain why probability weights matter:

• Each outcome of broken eggs, like 0 or 1, is multiplied by its probability (weight).
• These weighted values are combined to compute the expected value.

The reason \( \mu_y \) is not just the average of the numbers 0, 1, 2, 3, and 4 is because the probabilities, or weights, differ and are not equal. Most weights are skewed towards fewer broken eggs, reflecting that it’s more common to have 0 or 1 broken egg than 3 or 4. This impact of unequal weights results in an expected value of 0.56 rather than the simple average of 2.0, providing a more accurate measure that considers both value and frequency.