The Multinomial Distribution: Counting Outcomes Across Several Categories

The multinomial distribution extends the binomial idea from two outcomes to kk outcomes. In nn independent trials, each trial lands in exactly one of kk categories with fixed probabilities p1,p2,…,pkp_1, p_2, \dots, p_k (where p1+⋯+pk=1p_1+\cdots+p_k=1). If X1,…,XkX_1,\dots,X_k count how many trials fall into each category (so X1+⋯+Xk=nX_1+\cdots+X_k=n), the joint probability function is:


P(X1=x1,…,Xk=xk)=n!x1! x2!⋯xk! p1x1p2x2⋯pkxkP(X_1=x_1,\dots,X_k=x_k) = \frac{n!}{x_1!\,x_2!\cdots x_k!}\,p_1^{x_1}p_2^{x_2}\cdots p_k^{x_k}
Classification probabilities for each of n = 12 bottles inspected on a production line.

Classification probabilities for each of n = 12 bottles inspected on a production line.

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Marginal and conditional structure

  • Each individual XiX_i on its own is binomial: Xi∼Binomial(n,pi)X_i \sim \text{Binomial}(n,p_i), because we can lump every other category into "not category ii".
  • Given the total in two categories, say XA+XB=mX_A+X_B=m, the conditional distribution of XAX_A is Binomial ⁣(m,pApA+pB)\text{Binomial}\!\left(m,\dfrac{p_A}{p_A+p_B}\right), the other categories drop out because we already know the trials landed in AA or BB.

Worked example

A symmetric four-sided die is rolled n=5n=5 times, so each face has p=1/4p=1/4. Find P(2,1,1,1)P(2,1,1,1), i.e. face 1 appears twice and faces 2, 3, 4 appear once each.


5!2!1!1!1!(14)5=60×11024≈0.0586\frac{5!}{2!1!1!1!}\left(\tfrac14\right)^5 = 60 \times \frac{1}{1024} \approx 0.0586



The figure shows the classification probabilities used when inspecting bottles: Pass = 0.85, Minor defect = 0.10, Major defect = 0.05, independently across bottles. If n=12n = 12 bottles are inspected, what is the probability that exactly 10 are Pass, 1 is Minor defect, and 1 is Major defect?

Classification probabilities for each of n = 12 bottles inspected on a production line.

Classification probabilities for each of n = 12 bottles inspected on a production line.

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Choose an answer

Create a free account to get started

Unlock full course notes by signing up

  • Video lessons taught by top educators
  • Practice questions with guided solutions
  • Written review notes for reference
  • Expert answers anytime from course tutors

Which of the following experiments can be correctly modelled with a multinomial distribution?

Choose an answer

Create a free account to get started

Unlock full course notes by signing up

  • Video lessons taught by top educators
  • Practice questions with guided solutions
  • Written review notes for reference
  • Expert answers anytime from course tutors

Using the graph of the marginal distribution of the number of passing bottles (with n=12n = 12, p=0.85p = 0.85), what is the probability that at least 10 of the 12 bottles pass?

Classification probabilities for each of n = 12 bottles inspected on a production line.

Classification probabilities for each of n = 12 bottles inspected on a production line.

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Marginal Distribution of the Number of Passing Bottles (n = 12, p = 0.85)

Choose an answer

Create a free account to get started

Unlock full course notes by signing up

  • Video lessons taught by top educators
  • Practice questions with guided solutions
  • Written review notes for reference
  • Expert answers anytime from course tutors