Statistics How To

Uniform Distribution / Rectangular Distribution: What is it?

Probability and Statistics > Basic Graphs and Charts > Uniform Distribution

Contents:

  1. What is a Uniform Distribution?
  2. Expected Value/Mean and Variance.

1. What is a Uniform Distribution?

A uniform distribution, also called a rectangular distribution, is a probability distribution that has constant probability.

Uniform distribution

Image courtesy of the University of Houston.

This distribution is defined by two parameters, a and b:

  • a is the minimum.
  • b is the maximum.

The distribution is written as U(a,b).

The following graph shows the distribution with a=1 and b=3:

uniform distribution a b

Like all probability distributions for continuous random variables, the area under the graph of a random variable is always equal to 1. In the above graph, the area is:
A = l x h = 2 * 0.5 = 1.

This distribution can be broken down into two types. The most common type you’ll find in elementary statistics is the continuous uniform distribution (in the shape of a rectangle). However, there is a second type: the discrete uniform distribution. It still resembles a rectangle but instead of a line, a series of dots represent a known, finite number of outcomes. The following graph shows 5 possible outcomes:

Discrete uniform distribution. Image: ikamusumeFan|Wikimedia Commons

Image: ikamusumeFan|Wikimedia Commons

Rolling a single die is one example of a discrete uniform distribution; a die roll has four possible outcomes: 1,2,3,4,5, or 6. There is a 1/6 probability for each number being rolled.

The general formula for the probability density function (pdf) for the uniform distribution is:
f(x) = 1/ (B-A) for A≤x≤B.
“A” is the location parameter: The location parameter tells you where the center of the graph is. “B” is the scale parameter: The scale parameter stretches the graph out on the horizontal axis. Note: The A and B here aren’t to be confused with lowercase (a,b), which is an interval.

The uniform distribution doesn’t always look like a rectangle. A special case, the uniform cumulative distribution function, adds up all of the probabilities (in the same way a cumulative frequency distribution adds probabilities) and plots the result, which is a linear graph and not a rectangle:
uniform cdf

Variables

The variables in a uniform distribution are called uniform random variables.

2. Expected Value and Variance.

The expected value (i.e. the mean) of a uniform random variable X is:

E(X) = (1/2) (a + b)

This is also written equivalently as:
E(X) = (b + a) / 2.
“a” in the formula is the minimum value in the distribution, and “b” is the maximum value.
mean of uniform distribution


The variance of a uniform random variable is:

Var(x) = (1/12)(b-a)2

For the above image, the variance is (1/12)(3 – 1)2= 1/12 * 4 = 1/3.

3. Finding Probabilities for a Continuous Uniform Distribution

Sample question #1: The average amount of weight gained by a person over the winter months is uniformly distributed from 0 to 30lbs. Find the probability a person will gain between 10 and 15lbs during the winter months.

probability uniform distribution


Step 1: Find the height of the distribution. The area under a probability distribution is always 1. As there are 30 units (from zero to 30), then the height is 1/30.

Step 2: Find the width of the “slice” of the distribution mentioned in the question. Do this by subtracting the biggest number (b) from the smallest (a), to get b – a = 15 – 10 = 5.

Step 3: Multiply the width (Step 2) by the height (Step 1) to get:
Probability = 5 * 1/30 = 5/30 = 1/6.

Example Question 2: Find P(X≤10) for the above question.
This is asking you to find the probability that the random variable X is less than 10. In other words, you want to know the probability a person will gain up to ten pounds.

Step 1: Find the width of the “box”: b – a = 10 – 0 = 10.

Step 2: Multiply the width (Step 1) by the height. We already know the height is 1/30 (from example question 1), so:
10 * 1/30 = 10/30 = 1/3.

Example Question 3: Find P(20≤X≤25) for the above question. This is asking the probability of a weight gain between 20 and 25 pounds.

Step 1: Find the width of the “box”: b – a = 25 – 20 = 5.

Step 2: Multiply the width (Step 1) by the height. We already know the height is 1/30 (from example question 1), so:
5 * 1/30 = 5/30 = 1/6.

The More Formal Formula

You can solve these types of problems using the steps above, or you can us the formula for finding the probability for a continuous uniform distribution:

P(X) = d – c / b – a.

This is also sometimes written as:
P(X) = x2 – x1 / b – a.
“d” and “c” (x2 – x1)are the upper and lower bounds of the area you are trying to find.
You get exactly the same answer as if you’d followed the steps above. If formulas work for you…great. Personally, I find it easier to visualize these problems as trying to find an area inside a rectangle. Otherwise, I’ve just got another formula to memorize.

------------------------------------------------------------------------------

If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this Statistics with R track.

Comments are now closed for this post. Need help or want to post a correction? Please post a comment on our Facebook page and I'll do my best to help!
Uniform Distribution / Rectangular Distribution: What is it? was last modified: October 15th, 2017 by Stephanie Glen

5 thoughts on “Uniform Distribution / Rectangular Distribution: What is it?

  1. Mae Genato

    Can you give me a real life example with computation of a non-uniform discrete distribution

  2. Andale Post author

    Well, there’s dozens of different types of non-uniform distributions (i.e. normal, skewed, Student’s T). I’m not sure what you mean “with computation” though…computation for what? (i.e. the mean, a probability, future trends etc).

  3. CSD

    Looks like you are mistaken by writing the expected value formula as E(X) = (1/2) a + b
    It should be written as E(X)=(1/2)(a+b)