Actuarial Statistics · Jointly distributed random variables

Joint Distributions of Discrete and Continuous Variables

Q: What is the difference between a joint pmf and a joint cdf?

The joint pmf p(x, y) gives P(X = x, Y = y) for discrete variables, one pair at a time. The joint cdf F(x, y) gives P(X ≤ x, Y ≤ y), which adds all probability up to that point. The cdf exists for both discrete and continuous variables.

Q: How do I find probability from a joint pdf over a region?

Sketch the support and shade the event region. Then integrate f(x, y) over the shaded area, using limits where the inner variable depends on the outer one if the boundary is slanted. Evaluate the inner integral first.

Q: Can a joint pdf be greater than 1?

Yes. A pdf value is a density, not a probability. Only the volume under the surface over a region is a probability, and the total volume must equal 1.

Q: How do I find the constant in a joint pdf or pmf?

Set the total of the function over its whole support equal to 1. For a pdf, integrate over the support. For a pmf, add all the entries. Then solve for the constant.

Q: How do I get the joint pdf from the joint cdf?

Differentiate the cdf once with respect to each variable: f(x, y) = ∂²F ÷ ∂x∂y, wherever the derivative exists. Work separately in each region where F has a different formula.

Updated 11 October 2026 · Fact-checked

A joint distribution describes two random variables together. For discrete variables you use a joint pmf, p(x, y) = P(X = x, Y = y). For continuous variables you use a joint pdf f(x, y) and integrate it over a region to get probability. To solve questions, sketch the region, set the limits, then sum or integrate.

Understand Joint Distributions of Discrete and Continuous Variables

One random variable describes one quantity, such as a claim size. Often you need two together, such as claim size and claim count, or the lifetimes of two people. A joint distribution tells you how probability is shared across all pairs of values (x, y).

If X and Y are both discrete, the joint probability mass function is p(x, y) = P(X = x, Y = y). It is a table of probabilities. Every entry is at least 0 and all entries add up to 1. To find the probability of an event, add the entries for the pairs in that event.

If X and Y are both continuous, the joint probability density function f(x, y) is a surface. A single point has probability zero. Probability is the volume under the surface above a region A: P((X, Y) ∈ A) = ∫∫ f(x, y) dx dy over A. The total volume must be 1, and f(x, y) ≥ 0 everywhere.

The joint cumulative distribution function works for both types: F(x, y) = P(X ≤ x, Y ≤ y). For discrete variables you add p over all pairs with u ≤ x and v ≤ y. For continuous variables you integrate f over u ≤ x and v ≤ y. The pmf or pdf gives probability at or near a point. The cdf gives accumulated probability up to a point.

Some questions mix the two types: X discrete and Y continuous. The same idea holds. You sum over the discrete variable and integrate over the continuous one. The hardest part of most exam questions is not the calculus. It is getting the region and its limits right.

Key rules to remember

Joint pmf conditions: p(x, y) = P(X = x, Y = y); p(x, y) ≥ 0; Σ Σ p(x, y) = 1; The double sum runs over all possible pairs. Use this to find an unknown constant.
Joint pdf conditions: f(x, y) ≥ 0; ∫∫ f(x, y) dx dy = 1 over the whole plane; Integrate over the support only, where f is non-zero. Use this to find an unknown constant.
Probability over a region (discrete): P((X, Y) ∈ A) = Σ p(x, y) over pairs (x, y) in A; List the pairs in A, then add.
Probability over a region (continuous): P((X, Y) ∈ A) = ∫∫_A f(x, y) dx dy; The limits must describe A intersected with the support.
Joint cdf: F(x, y) = P(X ≤ x, Y ≤ y); Defined for discrete and continuous variables.
Joint cdf from joint pdf: F(x, y) = ∫ from -∞ to x ∫ from -∞ to y f(u, v) dv du; Only the part of the range inside the support contributes.
Joint pdf from joint cdf: f(x, y) = ∂²F(x, y) ÷ ∂x∂y; Holds where the derivative exists. For discrete variables, differences of F give p instead.
Rectangle probability from the cdf: P(a < X ≤ b, c < Y ≤ d) = F(b, d) − F(a, d) − F(b, c) + F(a, c); Add back F(a, c) because it is subtracted twice.

How to solve Joint Distributions of Discrete and Continuous Variables questions

Use this order for any question on joint pmfs, pdfs or cdfs. Most lost marks come from steps 2 and 3.

1Identify the type. Discrete means sum over pairs. Continuous means integrate. Mixed means sum over the discrete variable and integrate over the continuous one.
2Write down the support clearly, for example 0 < x < y < 1. Sketch it if the limits depend on each other.
3Mark the event region on the same sketch, for example X + Y < 1 or X < Y. The region you integrate over is the overlap of the event and the support.
4If a constant is unknown, set the total probability equal to 1 and solve for it first.
5Set up the limits. Choose the outer variable's range as constants and the inner variable's limits as functions of the outer one. Choose the order that makes the limits simplest.
6Evaluate the sum or integral carefully, inner first. Keep fractions exact.
7Check that the answer lies between 0 and 1. If you can, compute the complement as a cross-check.
8State the answer in words, with the notation used in the question.

Quickest way: Sketch, split, complement

When to use it: Use this when the region is awkward and time is short, especially for MCQs with a joint pdf on a triangle or square.

Sketch the support in 10 seconds. A rough triangle or square is enough.
Shade the event. If the shaded part is more complicated than the unshaded part, compute the complement and subtract from 1.
For a constant joint pdf on a region, probability is area of the event ÷ area of the support. No integral needed.
For a discrete table, circle the cells in the event and add them. Check the total of the table equals 1 first.
Choose the integration order that gives a single set of limits. If the region needs two pieces in one order, try the other order.

Common mistakes in Joint Distributions of Discrete and Continuous Variables

Integrating over the wrong region, such as the whole unit square when the event is X < Y.
Students skip the sketch and write limits from memory.
Fix: Always sketch the support and shade the event. Write limits only after reading them off the sketch.
Using constant limits for both variables when the support is a triangle, such as 0 < x < y < 1.
Students treat the variables as if they were independent.
Fix: The inner integral's limits must depend on the outer variable. The outer limits must be constants.
Forgetting to find the constant k first, or using a different value of k later.
Students rush to the probability asked for.
Fix: Set the total integral or sum equal to 1 as your first calculation and write k clearly.
Treating f(x, y) as a probability, or treating P(X = x, Y = y) as non-zero for continuous variables.
The pmf and pdf look similar in notation.
Fix: A pdf value can exceed 1. For continuous variables only regions have probability, and a single point has probability 0.
Confusing the joint pmf with the joint cdf, for example reading F(1, 1) as P(X = 1, Y = 1).
Both are written as functions of two values.
Fix: F(x, y) always uses ≤ for both variables, so it adds all cells up to and including x and y.
Using the cdf rectangle formula with only two terms, such as F(b, d) − F(a, c).
Students copy the one-variable result P(a < X ≤ b) = F(b) − F(a).
Fix: Use all four terms: F(b, d) − F(a, d) − F(b, c) + F(a, c).

Worked examples

Example 1

The continuous variables X and Y have joint pdf f(x, y) = k(x + 2y) for 0 < x < 1 and 0 < y < 1, and 0 otherwise. (a) Find k. (b) Find P(X < Y).

Show the solution

Total probability must be 1: ∫ from 0 to 1 ∫ from 0 to 1 k(x + 2y) dx dy = 1.
Inner integral over x: ∫ from 0 to 1 (x + 2y) dx = 1/2 + 2y.
Outer integral over y: ∫ from 0 to 1 (1/2 + 2y) dy = 1/2 + 1 = 3/2. So (3/2)k = 1 and k = 2/3.
For P(X < Y), the region inside the unit square is 0 < x < y, 0 < y < 1. It is the triangle above the line y = x.
Inner integral over x: ∫ from 0 to y (x + 2y) dx = y²/2 + 2y² = 5y²/2.
Outer integral: ∫ from 0 to 1 (5y²/2) dy = 5/6.
Multiply by k: P(X < Y) = (2/3) × (5/6) = 10/18 = 5/9.
Check: P(X > Y) = (2/3) ∫ from 0 to 1 ∫ from 0 to x (x + 2y) dy dx = (2/3) ∫ 2x² dx = (2/3)(2/3) = 4/9. And 5/9 + 4/9 = 1.

Answer: k = 2/3 and P(X < Y) = 5/9.

Example 2

The discrete variables X and Y have joint pmf p(x, y) = k(x + y) for x = 0, 1, 2 and y = 1, 2, and 0 otherwise. (a) Find k. (b) Find P(X + Y ≤ 3). (c) Find the joint cdf value F(1, 1).

Show the solution

List the values of x + y: (0,1) gives 1, (0,2) gives 2, (1,1) gives 2, (1,2) gives 3, (2,1) gives 3, (2,2) gives 4.
Sum = 1 + 2 + 2 + 3 + 3 + 4 = 15. Total probability is 1, so 15k = 1 and k = 1/15.
For P(X + Y ≤ 3), pick the pairs with x + y ≤ 3: (0,1), (0,2), (1,1), (1,2), (2,1). Their values are 1, 2, 2, 3, 3, which total 11.
So P(X + Y ≤ 3) = 11/15. Check by complement: only (2,2) is left, with probability 4/15, and 1 − 4/15 = 11/15.
F(1, 1) = P(X ≤ 1, Y ≤ 1). The pairs are (0,1) and (1,1) because y can only be 1 here.
F(1, 1) = 1/15 + 2/15 = 3/15 = 1/5.

Answer: k = 1/15, P(X + Y ≤ 3) = 11/15 and F(1, 1) = 1/5.

Exam tips

Draw the support and the event region in every continuous question, even in MCQs. Show the sketch or the limits in written answers, since method marks depend on correct limits.
Find any unknown constant first and state its value. A wrong k carries through every later part, so check it by recomputing the total.
Write the notation exactly: f(x, y) for the pdf, p(x, y) for the pmf, F(x, y) = P(X ≤ x, Y ≤ y) for the cdf. State the support each time.
Use a quick sanity check: a probability must lie between 0 and 1, and P(event) + P(complement) = 1. For constant pdfs, compare areas.
If a part asks for a marginal or conditional distribution next, keep your joint pdf and support clearly written. Those parts build directly on it.

Practice questions from Jointly distributed random variables