14.4.1. Conditional PMF

The conditional probability mass function (pmf) is defined as:

𝑃(𝑋 = 𝑥 ∣ 𝑌 = 𝑦) = \frac{𝑃(𝑋 = 𝑥,𝑌 = 𝑦)}{𝑃(𝑌 = 𝑦)}

The conditional pmf gives us the probability of a random variable, X, taking on a specific value x, under the condition that another random variable, Y, takes on a specific value y. This is expressed mathematically as the joint probability of both events occurring, divided by the probability of the given event Y. Essentially, it tells us how we can adjust our probability calculations when we know something about another random variable.

Imagine you're observing students in a class and you want to know the probability that a particular student scores a certain grade given that they studied for an exam. If you find out which students studied, you can tailor your expectations about their grades better than if you considered all students together, enhancing your understanding of how study habits affect performance.

The formula indicates these three components:

• 𝑃(𝑋 = 𝑥, 𝑌 = 𝑦): The probability that both X takes the specific value x and Y takes the specific value y.
• 𝑃(𝑌 = 𝑦): The total probability of Y being y.

This fraction effectively normalizes the joint probability, focusing it on the subset of outcomes where Y is known.

Here’s a breakdown: The numerator of the conditional pmf formula is the joint probability, which captures how likely it is that both X and Y occur together. The denominator adjusts this probability relative to the likelihood of Y itself occurring. This means that we are looking specifically at the scenarios where Y is y, allowing for a clearer understanding of how X behaves under this condition.

Think of a weather report that says there's a 70% chance of rain today. Now, if you find out that it's a cloudy day, the chances may shift - perhaps to 90%. The conditional pmf helps us understand these shifting probabilities when ‘conditions’ like 'it’s cloudy today' alter what we expect.

--