StatisticsProbability

Knowledge guide

How to Validate a Discrete Probability Distribution

Decide whether outcomes are discrete and verify a probability table by checking bounds, total probability, and complete outcome coverage.

Two separate questions

When a problem presents a table, do not merge these two checks:

  1. Is the random variable discrete?
  2. Is the listed distribution valid?

A variable can be discrete even when someone has written an invalid probability table for it.

What makes a variable discrete?

A discrete random variable has a finite or countably infinite set of possible values. Counts such as the number of messages, defects, or attempts are discrete. Measurements that can vary continuously over an interval, such as exact time or mass, are generally continuous.

Labels can also represent discrete categories when the probability model is explicitly defined over those categories, but numerical calculations such as a mean require meaningful numerical values.

The two probability rules

For every listed outcome x:

0 ≤ P(X=x) ≤ 1.

Across all mutually exclusive outcomes that cover the sample space:

ΣP(X=x) = 1.

Percentages must therefore total 100%. Small deviations may result from published rounding, but a solution should not silently assume rounding when the discrepancy is material or unexplained.

A fresh valid example

Let X be the number of messages received during a short interval.

x0123
P(X=x)0.100.350.400.15

The outcomes are countable, every probability is in [0,1], and

0.10 + 0.35 + 0.40 + 0.15 = 1.00.

Therefore this is a valid discrete probability distribution.

How invalid tables fail

A table is invalid if any probability is negative, any probability exceeds 1, the probabilities do not sum to 1, or the listed outcomes do not cover the intended sample space.

For example, replacing 0.15 above with 0.25 would make the total 1.10. Every individual entry would still look possible, but the table as a whole would be invalid.

A compact validation routine

  1. Define the random variable in words.
  2. Confirm that its possible outcomes are countable.
  3. Check every probability against [0,1].
  4. Add all probabilities using unrounded values when available.
  5. Confirm that the outcomes are mutually exclusive and exhaustive.

Only after all checks pass should you use the table to compute quantities such as E(X) = ΣxP(X=x).

Related question

Apply this knowledge

Use the concept guide to understand the reasoning, then return to the complete question and worked answer.

Determine Whether the Distribution Is a Discrete Probability Distribution: Household Size

Sources

These references support the core concepts and interpretation boundaries explained above.