WordGPT

Exponential distribution

Exponential distribution: the waiting time between events that occur at a steady rate. With the default parameters (rate λ = 1) its mean is 1 and its standard deviation 1. Use the simulator below to draw samples and see it take shape.

Simulate it

Change the parameters, then draw random samples. The bars are the samples; the line is the exact density. With more samples the bars settle onto it.

Samples to draw

00.20.40.60.810123456x
Mean
sample - · exact 1
Standard deviation
sample - · exact 1

What is the exponential distribution?

The exponential distribution describes the time you wait until the next event when events happen independently at a constant average rate λ. The time between customer arrivals, between phone calls or between radioactive decays is exponential, and so is the lifetime of a component that fails at a constant rate.

Like the geometric distribution it is memoryless: having already waited ten minutes tells you nothing about how much longer you will wait. It is the continuous-time partner of the Poisson distribution.

Where it is used

  • Time between customers arriving at a shop or calls at a call centre.
  • Lifetime of an electronic component with a constant failure rate.
  • Time between earthquakes in a region, approximately.
  • Call durations, approximately.
  • Time until a radioactive atom decays.

Shape and parameters

The density starts at its highest value λ at zero and falls steadily, with a long right tail. The mean and the standard deviation are both 1/λ, and the median is ln 2 / λ, about 0.69 of the mean.

What it assumes

  • Events occur independently.
  • The rate is constant, so the chance of an event in the next moment never changes.

How it relates to other distributions

  • Poisson distribution: Counts of the same events in a fixed interval are Poisson; the gaps between them are exponential.
  • Gamma distribution: The waiting time for the k-th event is gamma; an exponential is gamma with k = 1.
  • Geometric distribution: The discrete counterpart.
  • Weibull distribution: A Weibull with shape 1 is exponential; other shapes let the failure rate change with age.

Common mistakes

  • Using it for lifetimes where the failure rate rises with age (wear-out): use the Weibull.
  • Mixing up the rate λ with the mean 1/λ: software parameterises it in both ways.

How to work it out

  1. 1

    The probability density function (the formula that gives the shape):

    f(x)=λe−λx,x≥0f(x)=\lambda e^{-\lambda x},\quad x\ge 0
  2. 2

    Its mean (the long-run average) and variance (the spread):

    μ=1λ,σ2=1λ2\mu = \frac{1}{\lambda},\qquad \sigma^2 = \frac{1}{\lambda^2}
  3. 3

    Worked example with the default parameters (rate λ = 1): With rate λ = 1, what is the probability of waiting more than 2 time units?

    answer=0.1353\text{answer} = 0.1353
LaTeX source of the formulas on this page
\[ f(x)=\lambda e^{-\lambda x},\quad x\ge 0 \]
\[ \mu = \frac{1}{\lambda},\qquad \sigma^2 = \frac{1}{\lambda^2} \]
\[ \text{answer} = 0.1353 \]

At a glance

PropertyValue
TypeContinuous (measurements)
Parametersrate λ
Possible values0 to ∞
Mean1
Variance1
Standard deviation1
Example (rate λ = 1)With rate λ = 1, what is the probability of waiting more than 2 time units? Answer: 0.1353.

How to compute it in Excel, Python and R

ToolFunction
Excel / Google Sheets=EXPON.DIST(x, λ, TRUE) (FALSE for the density)
Python (SciPy)scipy.stats.expon.cdf(x, scale=1/lam)
Rpexp(x, rate = lambda)

Questions

What is the exponential distribution used for?

The waiting time between events that occur at a steady rate. Typical uses: Time between customers arriving at a shop or calls at a call centre. Lifetime of an electronic component with a constant failure rate. Time between earthquakes in a region, approximately.

What are the mean and variance of the exponential distribution?

Mean: μ = (1)/(λ). Variance: σ² = (1)/(λ²).

What does the exponential distribution assume?

Events occur independently. The rate is constant, so the chance of an event in the next moment never changes.

How do I calculate it in Excel?

=EXPON.DIST(x, λ, TRUE) (FALSE for the density)

How is the exponential distribution related to other distributions?

Poisson distribution: Counts of the same events in a fixed interval are Poisson; the gaps between them are exponential. Gamma distribution: The waiting time for the k-th event is gamma; an exponential is gamma with k = 1. Geometric distribution: The discrete counterpart. Weibull distribution: A Weibull with shape 1 is exponential; other shapes let the failure rate change with age.

Related distributions

Choosing a distribution

    Exponential distribution: what it is, when to use it, formulas and simulator | WordGPT