Contact

PROB in Apple Numbers: Probability of Discrete Values and Ranges

Statistical Intermediate Apple Numbers

Sum probabilities for exact discrete outcomes or inclusive intervals in Apple Numbers using aligned value and probability ranges.

Verification status: Official documentation referenced; example outcomes independently reasoned or arithmetically checked where stated. Not executed in Apple Numbers. How formula examples are checked

Quick answer. Given discrete outcomes 12, 13, 14, 15 and 16 with probabilities 0.17, 0.32, 0.15, 0.19 and 0.17, respectively, =PROB(A1:A5,B1:B5,15,16) returns 0.36 in Apple’s published example. The four-argument version includes both bounds. With the upper bound omitted, =PROB(A1:A5,B1:B5,14) returns 0.15 for the exact matching value 14. These are Apple’s documented expected results, not native results from this review. [A1]

When PROB is the right function

PROB sums already-known weights associated with a discrete set of numeric outcomes. It is useful for a survey-choice distribution, labelled game outcome, small demand scenario or discrete model whose weights are explicitly listed. It does not estimate those weights from a raw observation sample, assume a normal distribution or calculate a p-value for statistical significance. Your model must already provide probabilities that add up to 1. [A1, C1]

Mathematically, when each outcome xᵢ has weight pᵢ, P(lower ≤ X ≤ upper)=Σ pᵢ for all rows whose outcomes lie inside the inclusive interval. This defines an interval on the outcome values, not on row numbers or positions. [A1, C1]

Exact Apple syntax and arguments

=PROB(num-set, probability-values, lower, upper)
Position Apple argument Apple type Required? Meaning Documented default
1 num-set Collection of number values Yes Numeric outcome corresponding to each probability None
2 probability-values Collection containing number values; Apple says strings are ignored Yes Probability assigned to each outcome; numeric probabilities sum to 1 None
3 lower Number Yes Inclusive minimum outcome None
4 upper Number No Inclusive maximum outcome If omitted, select the value equal to lower

The two collections must have equal size, and the numeric probability values must sum to 1; Apple’s documentation says text values in collections are ignored. For reliable worksheet design, use only numeric outcome/probability pairs with no embedded string rows and verify the sum explicitly, rather than guessing how an individual text cell could affect pairwise alignment. Each probability should be a valid probability between 0 and 1 as a mathematical modelling requirement; Apple’s specific article only explicitly states that the numeric probabilities sum to 1, so this guide does not claim a precise negative-weight or rounding error behavior in Numbers. [A1, C1]

Build a validated probability table

Make a Numbers table named Choices with headers in B1 Choice and C1 Probability. Enter:

Row B (Choice) C (Probability)
2 12 0.17
3 13 0.32
4 14 0.15
5 15 0.19
6 16 0.17

In an empty result cell outside the input data, verify normalization with:

=SUM(C2:C6)

Expected: 1. Now use:

=PROB(B2:B6,C2:C6,15,16)

Expected: 0.36 (=0.19+0.17). The last argument is optional, so an exact-value question is:

=PROB(B2:B6,C2:C6,14)

Expected: 0.15. The ranges differ only by an offset from Apple’s no-header A1:A5/B1:B5 example; the numeric dataset and expected arithmetic are identical. This table layout uses Apple’s ordinary range references; the shifted version has not been executed in Numbers. [A1, A3, C1]

Advanced interval questions

Probability of a choice from 13 through 15 inclusive:

=PROB(B2:B6,C2:C6,13,15)

Expected 0.66 (=0.32+0.15+0.19). Both endpoints are counted. [A1, C1]

Probability of a choice below 15:

=PROB(B2:B6,C2:C6,12,14)

Expected 0.64 (=0.17+0.32+0.15). This illustrates a finite distribution whose minimum is known to be 12; PROB is not documented as having an unbounded-lower shortcut, so use a genuine minimum bound from the model rather than guessing a special wildcard. [A1, C1]

Probability of 15 or 16 by complement:

=1-PROB(B2:B6,C2:C6,12,14)

Expected 0.36, matching the direct bounded expression because the validated distribution totals 1. For a model where the complete list or normalization is uncertain, the complement may be misleading. [C1]

Scenario thresholds from cells. Store the lower and upper bounds in E2 and F2. Then enter =PROB(B2:B6,C2:C6,$E$2,$F$2) in a result cell, e.g., H2. The parameter cells can be changed without rewriting the input ranges. The $ characters fix only the parameter cells for copied formulas; the event ranges still need appropriate copying controls. [A3]

Troubleshooting and validity safeguards

Symptom Verify
Probability sum isn’t 1 Use =SUM(C2:C6) and audit numeric probabilities/formatting. Apple’s requirement is that their sum equals 1.
Unexpected output Confirm the outcome and probability ranges have the same dimensions, and probability for row 3 still corresponds to outcome in row 3.
Upper bound omitted Apple interprets this as an exact match to lower, not an open-ended cumulative probability.
Endpoints excluded Apple states ≥ lower and ≤ upper; endpoints should be included in the mathematical sum.
Text or blank cells in a collection Apple says strings are ignored. Do not assume another spreadsheet’s implicit conversion, or that ignored strings preserve the intended pairing in every malformed dataset. Clean the two numeric columns first.
Invalid weights Reject negative probabilities, weights above 1 or incomplete total as invalid probability models, even if an application might attempt a numerical sum. Precise Numbers error messages are not documented here.
Locale punctuation mismatch Apple’s English comma syntax becomes a semicolon-separated formula when the decimal separator is a comma.
Range mistaken for rows PROB filters numeric outcome values, not row numbers or percentage ranks.

No precise error token (#VALUE!, etc.), floating-point sum tolerance, duplicate-outcome aggregation rule or automatic text-coercion behavior is asserted without native evidence. Also distinguish a displayed percentage from the stored decimal (36% = 0.36). [A1, A2, C1]

Numbers version, table references, and locale

These instructions follow Apple’s English Formulas and Functions Help, version 15.4, reviewed 13 October 2026. Apple describes the Functions Browser for Numbers on Mac and on iPhone/iPad, and in relevant iWork apps on iCloud.com, but the documentation version is not proof that every function exists in every historical app release or platform build. There was no native Numbers execution for this guide. Do not represent the expected outputs as measured application results. [A1, A4]

The examples use English function names, a period as decimal point and a comma between arguments. Apple explicitly says that a Numbers setup using a comma for decimals requires a semicolon between function arguments when typing formulas. Spreadsheet language/region settings also affect numeric punctuation; select the appropriate localized function spelling in your own Functions Browser rather than creating a second English-site URL from a translated name. [A2, A5]

Inside a Numbers table, B2:C6 identifies the rectangular range and $D$2 is an absolute reference. A cross-table reference can be written as Table Name::B2 or Sheet Name::Table Name::B2 where required, and Numbers can insert the appropriate reference when you select a source cell. The formulas below assume all inputs and the result are in the same named table; do not copy an ambiguous short reference across multiple tables. [A3]

Scope of verification

  • Official documentation check: Apple’s named function help, argument/type and cell-reference instructions, and the functions-overview regional-syntax rule were read on 13 October 2026. The source links and claim map are preserved below.
  • Independent calculations: Expected numbers were recomputed with Python/SciPy or exact arithmetic; these checks establish the mathematics and consistency with Apple’s published examples, not the installed application’s behaviour.
  • Not performed: native Apple Numbers execution, screenshots or historical-platform regression tests.

Official references checked (13 October 2026)