DataTerminology.comNEW
A probability distribution defined over a finite or countably infinite set of outcomes via a probability mass function.
DESCRIPTION
Discrete distributions model random variables that take on separate, countable outcomes. Common examples include Bernoulli, Binomial, Poisson, and Geometric distributions. They are described by probability mass functions, with sums over outcomes equal to one. Applications span event counts, successes in trials, and arrival processes. Choosing an appropriate distribution improves modeling accuracy and inference reliability.
ALTERNATIVE DEFINITIONS
Wikipedia
A discrete probability distribution is the probability distribution of a discrete random variable, i.e., one that takes on a countable number of distinct values.
Clarifies countable outcomes as the defining feature.
Encyclopaedia Britannica
A probability distribution can be discrete, assigning probabilities to distinct outcomes, or continuous, assigning probabilities over intervals.
Contrasts discrete and continuous distributions.
NIST/SEMATECH e-Handbook of Statistical Methods
Discrete distributions are characterized by a probability mass function that assigns probabilities to individual, countable outcomes.
Centers on the probability mass function characterization.
Penn State Eberly College of Science
A discrete distribution specifies the probabilities of a discrete random variable’s possible values, often summarized by a probability mass function.
Emphasizes PMF as summary of probabilities.
RELATED TERMS
Discrete Random Variable Probability Mass Function Bernoulli Distribution Binomial Distribution Poisson Distribution Geometric Distribution Cumulative Distribution Function Expectation Variance Law of Total Probability
COMPETING TERMS
KNOWLEDGE AREAS
Mathematics Data Science Testing & Validation
TARGET AUDIENCE
Data Scientist Data Analyst AI Engineer
EXTERNAL SOURCES
- Wikipedia - Discrete probability distribution
- NIST - NIST overview of discrete distributions
- Penn State - Course notes on discrete distributions
- Khan Academy - Probability library with discrete topics
TOP QUESTIONS
How do I identify a Discrete Distribution in data?
Look for countable outcomes (e.g., counts, categories) and model with a PMF rather than a density.
Category: Guidance
What’s the difference between a Discrete Distribution and a continuous one?
Discrete assigns probabilities to exact outcomes; continuous assigns probability to intervals via densities.
Category: Comparison
Which models commonly use Discrete Distributions?
Count models (Poisson, Negative Binomial) and binomial processes are typical examples.
Category: Example
What’s a common mistake with Discrete Distributions?
Using a normal approximation when counts are small or skewed can misrepresent tail probabilities.
Category: Mistakes
How to estimate parameters of a Discrete Distribution?
Use MLE, method of moments, or Bayesian inference depending on context and sample size.
Category: Requirements
What are risks of mis-specifying a Discrete Distribution?
Prediction errors and biased inferences, especially in the tails or rare-event modeling.
Category: Risk
How does the PMF define a Discrete Distribution?
It lists probability for each outcome and sums to one across all possible values.
Category: Concept
When is Poisson a good Discrete Distribution choice?
For counts of rare independent events over a fixed interval with constant rate.
Category: Guidance
How to visualize a Discrete Distribution?
Use bar charts of probabilities or counts and overlay theoretical PMF values.
Category: Application